• Europe’s $3.1B satellite merger won’t rival Musk’s Starlink

    Two satellite heavyweights are about to form a European rival to Starlink. But they’ll face an uphill battle to compete with Elon Musk’s firm.
    Luxembourg-based SES’ proposed bn takeover of Intelsat is set to get the green light from EU officials, Reuters reports. A final verdict is expected by June 10.  
    SES first announced plans to acquire Luxembourgish-American rival Intelsat in April 2024, calling it a “transformational merger” that could reshape the satellite internet market.
    The merged company would have a fleet of more than 100 geostationary and 26 medium Earth orbit satellites. Intelsat would contribute 75 of those probes, which provide a range of services, including TV, radio, satellite internet, and secure communications for governments and militaries.
    A SES-Intelsat tie-up would create Europe’s second-largest satellite internet provider, after Franco-British firm Eutelsat. Smaller contenders include the UK’s Inmarsat and Spain’s Hisdesat. 
    Combined, Europe’s satellite firms could offer the continent much-sought alternatives to Musk’s Starlink or Amazon’s Project Kuiper, at a time when tech sovereignty is high on the political agenda. 
    European leaders are increasingly concerned about relying on Starlink. Their fears have been stoked by reports that US officials threatened to cut off the system in Ukraine if the country didn’t meet their demands on sharing its mineral wealth.  
    The situation raised doubts about the security implications of Ukraine — and broader Europe — relying on a single, privately owned network whose boss has direct ties to the Trump administration. Eustalstat, SES, Inmarsat, and Hisdesat reportedly entered talks with EU governments in March about providing back-up connectivity to Ukraine. 
    Despite the SES–Intelsat merger creating a more formidable European player, the new alliance will still face a daunting challenge against Starlink’s dominance. 
    Musk’s firm dwarfs its competitors in satellite numbers, with over 7,000 in low-Earth orbit. Its closest rival, Eutelsat, has around 600. Meanwhile, Amazon’s Project Kuiper is planning to deploy a constellation of 3,236 satellites.  
    Being part of SpaceX also affords Starlink direct access to satellite manufacturing and launch capacity. In contrast, SES and Intelsat depend on third-party launch providers and currently lack a proprietary low-Earth orbit network. While Intelsat has a mn agreement to access Eutelsat’s LEO capacity, this reliance on external constellations puts the merged entity at a disadvantage. 
    Even when combined with Europe’s other players — including Eutelsat, Inmarsat, and Hisdesat — the proposed merger pales in comparison. As European policymakers push for strategic autonomy in space, the SES–Intelsat merger may be a step in the right direction, but it’s still a far cry from a true Starlink rival.
    #europes #31b #satellite #merger #wont
    Europe’s $3.1B satellite merger won’t rival Musk’s Starlink
    Two satellite heavyweights are about to form a European rival to Starlink. But they’ll face an uphill battle to compete with Elon Musk’s firm. Luxembourg-based SES’ proposed bn takeover of Intelsat is set to get the green light from EU officials, Reuters reports. A final verdict is expected by June 10.   SES first announced plans to acquire Luxembourgish-American rival Intelsat in April 2024, calling it a “transformational merger” that could reshape the satellite internet market. The merged company would have a fleet of more than 100 geostationary and 26 medium Earth orbit satellites. Intelsat would contribute 75 of those probes, which provide a range of services, including TV, radio, satellite internet, and secure communications for governments and militaries. A SES-Intelsat tie-up would create Europe’s second-largest satellite internet provider, after Franco-British firm Eutelsat. Smaller contenders include the UK’s Inmarsat and Spain’s Hisdesat.  Combined, Europe’s satellite firms could offer the continent much-sought alternatives to Musk’s Starlink or Amazon’s Project Kuiper, at a time when tech sovereignty is high on the political agenda.  European leaders are increasingly concerned about relying on Starlink. Their fears have been stoked by reports that US officials threatened to cut off the system in Ukraine if the country didn’t meet their demands on sharing its mineral wealth.   The situation raised doubts about the security implications of Ukraine — and broader Europe — relying on a single, privately owned network whose boss has direct ties to the Trump administration. Eustalstat, SES, Inmarsat, and Hisdesat reportedly entered talks with EU governments in March about providing back-up connectivity to Ukraine.  Despite the SES–Intelsat merger creating a more formidable European player, the new alliance will still face a daunting challenge against Starlink’s dominance.  Musk’s firm dwarfs its competitors in satellite numbers, with over 7,000 in low-Earth orbit. Its closest rival, Eutelsat, has around 600. Meanwhile, Amazon’s Project Kuiper is planning to deploy a constellation of 3,236 satellites.   Being part of SpaceX also affords Starlink direct access to satellite manufacturing and launch capacity. In contrast, SES and Intelsat depend on third-party launch providers and currently lack a proprietary low-Earth orbit network. While Intelsat has a mn agreement to access Eutelsat’s LEO capacity, this reliance on external constellations puts the merged entity at a disadvantage.  Even when combined with Europe’s other players — including Eutelsat, Inmarsat, and Hisdesat — the proposed merger pales in comparison. As European policymakers push for strategic autonomy in space, the SES–Intelsat merger may be a step in the right direction, but it’s still a far cry from a true Starlink rival. #europes #31b #satellite #merger #wont
    THENEXTWEB.COM
    Europe’s $3.1B satellite merger won’t rival Musk’s Starlink
    Two satellite heavyweights are about to form a European rival to Starlink. But they’ll face an uphill battle to compete with Elon Musk’s firm. Luxembourg-based SES’ proposed $3.1bn takeover of Intelsat is set to get the green light from EU officials, Reuters reports. A final verdict is expected by June 10.   SES first announced plans to acquire Luxembourgish-American rival Intelsat in April 2024, calling it a “transformational merger” that could reshape the satellite internet market. The merged company would have a fleet of more than 100 geostationary and 26 medium Earth orbit satellites. Intelsat would contribute 75 of those probes, which provide a range of services, including TV, radio, satellite internet, and secure communications for governments and militaries. A SES-Intelsat tie-up would create Europe’s second-largest satellite internet provider, after Franco-British firm Eutelsat. Smaller contenders include the UK’s Inmarsat and Spain’s Hisdesat.  Combined, Europe’s satellite firms could offer the continent much-sought alternatives to Musk’s Starlink or Amazon’s Project Kuiper, at a time when tech sovereignty is high on the political agenda.  European leaders are increasingly concerned about relying on Starlink. Their fears have been stoked by reports that US officials threatened to cut off the system in Ukraine if the country didn’t meet their demands on sharing its mineral wealth.   The situation raised doubts about the security implications of Ukraine — and broader Europe — relying on a single, privately owned network whose boss has direct ties to the Trump administration. Eustalstat, SES, Inmarsat, and Hisdesat reportedly entered talks with EU governments in March about providing back-up connectivity to Ukraine.  Despite the SES–Intelsat merger creating a more formidable European player, the new alliance will still face a daunting challenge against Starlink’s dominance.  Musk’s firm dwarfs its competitors in satellite numbers, with over 7,000 in low-Earth orbit (LEO). Its closest rival, Eutelsat, has around 600. Meanwhile, Amazon’s Project Kuiper is planning to deploy a constellation of 3,236 satellites.   Being part of SpaceX also affords Starlink direct access to satellite manufacturing and launch capacity. In contrast, SES and Intelsat depend on third-party launch providers and currently lack a proprietary low-Earth orbit network. While Intelsat has a $250mn agreement to access Eutelsat’s LEO capacity, this reliance on external constellations puts the merged entity at a disadvantage.  Even when combined with Europe’s other players — including Eutelsat, Inmarsat, and Hisdesat — the proposed merger pales in comparison. As European policymakers push for strategic autonomy in space, the SES–Intelsat merger may be a step in the right direction, but it’s still a far cry from a true Starlink rival.
    Like
    Love
    Wow
    Angry
    Sad
    220
    0 Comentários 0 Compartilhamentos 0 Anterior
  • The Resilient House Multi-Generational Housing / etal.

    The Resilient House Multi-Generational Housing / etal.this picture!© Federico FarinattiSocial Housing, Residential•München, Germany

    Architects:
    etal.
    Area
    Area of this architecture project

    Area: 
    930 m²

    Year
    Completion year of this architecture project

    Year: 

    2024

    Photographs

    Photographs:Federico Farinatti

    Manufacturers
    Brands with products used in this architecture project

    Manufacturers:  Persiana Barcelona More SpecsLess Specs
    this picture!
    Text description provided by the architects. The multi-generational house with communal forms of living was realized for and with a "Baugruppe"in Munich, with the aim of providing long-term affordable, self-managed rental housing. Through a concept selection process, the group was selected by the city to lease the land for the next 80 years. This marks the first new construction project of the Mietshäuser Syndikat in Munich. The building is a socially subsidized residential project under the "München Modell-Genossenschaften" funding model. The cooperative planning process was facilitated by the architectural office etal. All decisions by the group were reached through a consensus-based approach.this picture!this picture!The plot, located in the southeastern part of Munich, is situated in a predominantly residential area characterized by single-family homes and terraced houses. The three-story, barrier-free building accommodates one cluster apartment on each floor, with a communal living and dining area. Each individual housing unit is equipped with its own bathroom and the possibility of a kitchenette connection. On the ground floor, a multifunctional room serves both the building's residents and the local community as a space for multifunctional purposes. The basement contains further shared spaces, including a bicycle storage area, a wood workshop, and a laundry room. The building presents a three-story facade towards the street, while the garden side features a mansard roof forming a two-story facade. The remaining roof area is extensively greened and accommodates photovoltaic panels. The building was designed as a timber construction. All above-ground floors are constructed using timber frame construction, while the elevator shafts and ceilings are made from glued laminated timber.this picture!this picture!this picture!The roof is designed as an open rafter structure to make it visible throughout the residential spaces of the top floor. The vertically, story-wise stacked cladding made of local spruce and steel trapezoidal sheets as weather protection for the exterior wooden sunshading defines the suspended facade. To reduce costs, the external walls, insulated with cellulose and wood wool, were constructed without an additional shell for electrical wiring. The cement screed flooring was also left visible, only being sanded and oiled. The group's desire for individual rooms of approximately equal size deeply influenced the design concept. The distinctive yet simple structure offers long-term flexibility for various forms of living. Seven rooms, each approximately 18 sqm in size, are arranged around a central hallway and a bathroom core. The water connections are positioned in such a way that kitchens can be flexibly installed in six of the rooms without significant modifications.this picture!this picture!this picture!this picture!The walls of the rooms are designed as partition walls between apartments, providing the necessary soundproofing for possible reorganizations. So-called "breakpoints" consisting of lintels and thresholds allow for the addition or removal of rooms. These elements also make the potential for alterations visible to the residents. Functional elements, such as the wooden sunshading, allow residents to carry out alterations, maintenance, or repairs themselves through simple construction and installation techniques. The participatory process and the high level of self-involvement during construction strengthened the residents' identification with their house, ensuring that the knowledge about the building's adaptability remains accessible to future generations.this picture!

    Project gallerySee allShow less
    Project locationAddress:Görzer Straße 128, 8154, Munich, GermanyLocation to be used only as a reference. It could indicate city/country but not exact address.About this officeetal.Office•••
    MaterialsWoodSteelMaterials and TagsPublished on May 24, 2025Cite: "The Resilient House Multi-Generational Housing / etal." 24 May 2025. ArchDaily. Accessed . < ISSN 0719-8884Save世界上最受欢迎的建筑网站现已推出你的母语版本!想浏览ArchDaily中国吗?是否
    You've started following your first account!Did you know?You'll now receive updates based on what you follow! Personalize your stream and start following your favorite authors, offices and users.Go to my stream
    #resilient #house #multigenerational #housing #etal
    The Resilient House Multi-Generational Housing / etal.
    The Resilient House Multi-Generational Housing / etal.this picture!© Federico FarinattiSocial Housing, Residential•München, Germany Architects: etal. Area Area of this architecture project Area:  930 m² Year Completion year of this architecture project Year:  2024 Photographs Photographs:Federico Farinatti Manufacturers Brands with products used in this architecture project Manufacturers:  Persiana Barcelona More SpecsLess Specs this picture! Text description provided by the architects. The multi-generational house with communal forms of living was realized for and with a "Baugruppe"in Munich, with the aim of providing long-term affordable, self-managed rental housing. Through a concept selection process, the group was selected by the city to lease the land for the next 80 years. This marks the first new construction project of the Mietshäuser Syndikat in Munich. The building is a socially subsidized residential project under the "München Modell-Genossenschaften" funding model. The cooperative planning process was facilitated by the architectural office etal. All decisions by the group were reached through a consensus-based approach.this picture!this picture!The plot, located in the southeastern part of Munich, is situated in a predominantly residential area characterized by single-family homes and terraced houses. The three-story, barrier-free building accommodates one cluster apartment on each floor, with a communal living and dining area. Each individual housing unit is equipped with its own bathroom and the possibility of a kitchenette connection. On the ground floor, a multifunctional room serves both the building's residents and the local community as a space for multifunctional purposes. The basement contains further shared spaces, including a bicycle storage area, a wood workshop, and a laundry room. The building presents a three-story facade towards the street, while the garden side features a mansard roof forming a two-story facade. The remaining roof area is extensively greened and accommodates photovoltaic panels. The building was designed as a timber construction. All above-ground floors are constructed using timber frame construction, while the elevator shafts and ceilings are made from glued laminated timber.this picture!this picture!this picture!The roof is designed as an open rafter structure to make it visible throughout the residential spaces of the top floor. The vertically, story-wise stacked cladding made of local spruce and steel trapezoidal sheets as weather protection for the exterior wooden sunshading defines the suspended facade. To reduce costs, the external walls, insulated with cellulose and wood wool, were constructed without an additional shell for electrical wiring. The cement screed flooring was also left visible, only being sanded and oiled. The group's desire for individual rooms of approximately equal size deeply influenced the design concept. The distinctive yet simple structure offers long-term flexibility for various forms of living. Seven rooms, each approximately 18 sqm in size, are arranged around a central hallway and a bathroom core. The water connections are positioned in such a way that kitchens can be flexibly installed in six of the rooms without significant modifications.this picture!this picture!this picture!this picture!The walls of the rooms are designed as partition walls between apartments, providing the necessary soundproofing for possible reorganizations. So-called "breakpoints" consisting of lintels and thresholds allow for the addition or removal of rooms. These elements also make the potential for alterations visible to the residents. Functional elements, such as the wooden sunshading, allow residents to carry out alterations, maintenance, or repairs themselves through simple construction and installation techniques. The participatory process and the high level of self-involvement during construction strengthened the residents' identification with their house, ensuring that the knowledge about the building's adaptability remains accessible to future generations.this picture! Project gallerySee allShow less Project locationAddress:Görzer Straße 128, 8154, Munich, GermanyLocation to be used only as a reference. It could indicate city/country but not exact address.About this officeetal.Office••• MaterialsWoodSteelMaterials and TagsPublished on May 24, 2025Cite: "The Resilient House Multi-Generational Housing / etal." 24 May 2025. ArchDaily. Accessed . < ISSN 0719-8884Save世界上最受欢迎的建筑网站现已推出你的母语版本!想浏览ArchDaily中国吗?是否 You've started following your first account!Did you know?You'll now receive updates based on what you follow! Personalize your stream and start following your favorite authors, offices and users.Go to my stream #resilient #house #multigenerational #housing #etal
    WWW.ARCHDAILY.COM
    The Resilient House Multi-Generational Housing / etal.
    The Resilient House Multi-Generational Housing / etal.Save this picture!© Federico FarinattiSocial Housing, Residential•München, Germany Architects: etal. Area Area of this architecture project Area:  930 m² Year Completion year of this architecture project Year:  2024 Photographs Photographs:Federico Farinatti Manufacturers Brands with products used in this architecture project Manufacturers:  Persiana Barcelona More SpecsLess Specs Save this picture! Text description provided by the architects. The multi-generational house with communal forms of living was realized for and with a "Baugruppe" (a group of people who commission their own housing development, here: together with the "Mietshäuser Syndikat"- https://www.syndikat.org) in Munich, with the aim of providing long-term affordable, self-managed rental housing. Through a concept selection process, the group was selected by the city to lease the land for the next 80 years. This marks the first new construction project of the Mietshäuser Syndikat in Munich. The building is a socially subsidized residential project under the "München Modell-Genossenschaften" funding model. The cooperative planning process was facilitated by the architectural office etal. All decisions by the group were reached through a consensus-based approach.Save this picture!Save this picture!The plot, located in the southeastern part of Munich, is situated in a predominantly residential area characterized by single-family homes and terraced houses. The three-story, barrier-free building accommodates one cluster apartment on each floor, with a communal living and dining area. Each individual housing unit is equipped with its own bathroom and the possibility of a kitchenette connection. On the ground floor, a multifunctional room serves both the building's residents and the local community as a space for multifunctional purposes. The basement contains further shared spaces, including a bicycle storage area, a wood workshop, and a laundry room. The building presents a three-story facade towards the street, while the garden side features a mansard roof forming a two-story facade. The remaining roof area is extensively greened and accommodates photovoltaic panels. The building was designed as a timber construction. All above-ground floors are constructed using timber frame construction, while the elevator shafts and ceilings are made from glued laminated timber.Save this picture!Save this picture!Save this picture!The roof is designed as an open rafter structure to make it visible throughout the residential spaces of the top floor. The vertically, story-wise stacked cladding made of local spruce and steel trapezoidal sheets as weather protection for the exterior wooden sunshading defines the suspended facade. To reduce costs, the external walls, insulated with cellulose and wood wool, were constructed without an additional shell for electrical wiring. The cement screed flooring was also left visible, only being sanded and oiled. The group's desire for individual rooms of approximately equal size deeply influenced the design concept. The distinctive yet simple structure offers long-term flexibility for various forms of living. Seven rooms, each approximately 18 sqm in size, are arranged around a central hallway and a bathroom core. The water connections are positioned in such a way that kitchens can be flexibly installed in six of the rooms without significant modifications.Save this picture!Save this picture!Save this picture!Save this picture!The walls of the rooms are designed as partition walls between apartments, providing the necessary soundproofing for possible reorganizations. So-called "breakpoints" consisting of lintels and thresholds allow for the addition or removal of rooms. These elements also make the potential for alterations visible to the residents. Functional elements, such as the wooden sunshading, allow residents to carry out alterations, maintenance, or repairs themselves through simple construction and installation techniques. The participatory process and the high level of self-involvement during construction strengthened the residents' identification with their house, ensuring that the knowledge about the building's adaptability remains accessible to future generations.Save this picture! Project gallerySee allShow less Project locationAddress:Görzer Straße 128, 8154, Munich, GermanyLocation to be used only as a reference. It could indicate city/country but not exact address.About this officeetal.Office••• MaterialsWoodSteelMaterials and TagsPublished on May 24, 2025Cite: "The Resilient House Multi-Generational Housing / etal." 24 May 2025. ArchDaily. Accessed . <https://www.archdaily.com/1030454/the-resilient-house-multi-generational-housing-on-gorzer-street-etal&gt ISSN 0719-8884Save世界上最受欢迎的建筑网站现已推出你的母语版本!想浏览ArchDaily中国吗?是否 You've started following your first account!Did you know?You'll now receive updates based on what you follow! Personalize your stream and start following your favorite authors, offices and users.Go to my stream
    0 Comentários 0 Compartilhamentos 0 Anterior
  • Is Intel’s Arc Pro B60 the Dual GPU Innovation We’ve Been Waiting For?

    Key Takeaways

    Maxsun unveiled the new Intel Arc Pro B60 Dual GPU video card, featuring two 24GB Pro B60 GPUs.
    It features a blower-style fan, ideal for building complex workstations with multiple cards.
    Although the pricing isn’t made public yet, we expect it to be around which can be a competition-killer.

    Remember the good old CrossFire days when you could hook up multiple AMD GPUs to achieve higher performance? Well, those days might be returning thanks to the new Intel Arc Pro B60 Dual GPU launched by Maxsun at Computex.
    It PCIe features two 24GB GPUs in a single card, meaning you can get 48 GB of GPU memory. What’s incredible here is that the Arc Pro B60 only needs 8 PCIe lanes, unlike some high-end GPUs that need 16.

    However, you’ll need an X16 motherboard/CPU combo that supports PCIe bifurcation. In that case, the entire x16 slot can be split into two x8 connections.
    Interestingly, the Arc Pro B60 isn’t marketed as a gaming GPU – it’s meant primarily for workstations that require high computing power for tasks such as rendering, animations, moderate AI development, 3D modelling, etc.
    This is why it features a unique blower-like fan instead of the typical open-air fan we see in gaming CPUs. Unlike open-air fans that spread the heat out inside the case, a blower-style fan works like a hairdryer by pushing the heat out from the inside of the case.
    While the cooling efficiency of a blower fan might be lower on a single GPU, they turn out to be more effective in stacks, which is exactly what the Arc Pro B60 is meant for.
    So, if you’re building a workstation with 3 Arc Pro B60s, a blower-style fan ensures all 6 GPUs can be cooled without overwhelming the entire setup, making it better for ‘stacking.’
    Comeback of Multi-GPU Setups
    Multi-GPU setups were quite the thing back in the day. We’re sure that a couple of die-hard builders in our team would serve up puppy eyes thinking about hooking up two 16GB GPUs to get more power.
    However, despite the widespread belief that you would get double the performance, memory was only enhanced by 30-50%. Still, these CrossFire setups served the purpose, i.e., gamers could play resource-intensive games.
    It’s worth noting that such dual-GPU setups faded away with the rapid development of newer GPUs. In short, single GPUs became powerful enough to overshadow dual GPUs. Eventually, a lot of applications and games stopped supporting CrossFire setups.
    With the new Intel Arc Pro B60, dual GPUs might make a mini-comeback. They’re not returning to the gaming arena, but to data centers, workstations, and AI setups. NVIDIA hasn’t promoted this in a long time, pushing its expensive GPUs into the market instead.

    This is also precisely where Intel wants to punch harder – the price point. Although we’re still waiting for the exact details on the Intel Arc Pro B60 dual GPU setup price, experts believe it could be priced around -To get the same 48 GB of VRAM, you might have to spend -6,000 with models like the NVIDIA RTX 6000 Ada.
    Of course, there will be performance differences between the two, too, but getting at-par VRAM is the first step of the process. Honestly, if power-intensive workspaces are able to get 48 GB of VRAM at 70-80% less cost, this dual GPU setup can become a market killer.
    In other news, Intel has also announced Project Battlematrix – a solution that supports AI workloads that can combine up to eight Intel Arc Pro GPUs in a system, with 192 GB of VRAM.
    The company seems to be working hard on the workstation and data processing industry – something NVIDIA and AMD haven’t catered to explicitly yet. This could give Intel a first-mover advantage.
    If it can achieve comparable performance with some modern NVIDIA chips, the new Intel Arc Pro B60 can revive the Blue team.
    : Nvidia’s downgraded H20 chips might not be enough to stop China’s Ai ambitions

    Krishi is a seasoned tech journalist with over four years of experience writing about PC hardware, consumer technology, and artificial intelligence.  Clarity and accessibility are at the core of Krishi’s writing style.
    He believes technology writing should empower readers—not confuse them—and he’s committed to ensuring his content is always easy to understand without sacrificing accuracy or depth.
    Over the years, Krishi has contributed to some of the most reputable names in the industry, including Techopedia, TechRadar, and Tom’s Guide. A man of many talents, Krishi has also proven his mettle as a crypto writer, tackling complex topics with both ease and zeal. His work spans various formats—from in-depth explainers and news coverage to feature pieces and buying guides. 
    Behind the scenes, Krishi operates from a dual-monitor setupthat’s always buzzing with news feeds, technical documentation, and research notes, as well as the occasional gaming sessions that keep him fresh. 
    Krishi thrives on staying current, always ready to dive into the latest announcements, industry shifts, and their far-reaching impacts.  When he's not deep into research on the latest PC hardware news, Krishi would love to chat with you about day trading and the financial markets—oh! And cricket, as well.

    View all articles by Krishi Chowdhary

    Our editorial process

    The Tech Report editorial policy is centered on providing helpful, accurate content that offers real value to our readers. We only work with experienced writers who have specific knowledge in the topics they cover, including latest developments in technology, online privacy, cryptocurrencies, software, and more. Our editorial policy ensures that each topic is researched and curated by our in-house editors. We maintain rigorous journalistic standards, and every article is 100% written by real authors.
    #intels #arc #pro #b60 #dual
    Is Intel’s Arc Pro B60 the Dual GPU Innovation We’ve Been Waiting For?
    Key Takeaways Maxsun unveiled the new Intel Arc Pro B60 Dual GPU video card, featuring two 24GB Pro B60 GPUs. It features a blower-style fan, ideal for building complex workstations with multiple cards. Although the pricing isn’t made public yet, we expect it to be around which can be a competition-killer. Remember the good old CrossFire days when you could hook up multiple AMD GPUs to achieve higher performance? Well, those days might be returning thanks to the new Intel Arc Pro B60 Dual GPU launched by Maxsun at Computex. It PCIe features two 24GB GPUs in a single card, meaning you can get 48 GB of GPU memory. What’s incredible here is that the Arc Pro B60 only needs 8 PCIe lanes, unlike some high-end GPUs that need 16. However, you’ll need an X16 motherboard/CPU combo that supports PCIe bifurcation. In that case, the entire x16 slot can be split into two x8 connections. Interestingly, the Arc Pro B60 isn’t marketed as a gaming GPU – it’s meant primarily for workstations that require high computing power for tasks such as rendering, animations, moderate AI development, 3D modelling, etc. This is why it features a unique blower-like fan instead of the typical open-air fan we see in gaming CPUs. Unlike open-air fans that spread the heat out inside the case, a blower-style fan works like a hairdryer by pushing the heat out from the inside of the case. While the cooling efficiency of a blower fan might be lower on a single GPU, they turn out to be more effective in stacks, which is exactly what the Arc Pro B60 is meant for. So, if you’re building a workstation with 3 Arc Pro B60s, a blower-style fan ensures all 6 GPUs can be cooled without overwhelming the entire setup, making it better for ‘stacking.’ Comeback of Multi-GPU Setups Multi-GPU setups were quite the thing back in the day. We’re sure that a couple of die-hard builders in our team would serve up puppy eyes thinking about hooking up two 16GB GPUs to get more power. However, despite the widespread belief that you would get double the performance, memory was only enhanced by 30-50%. Still, these CrossFire setups served the purpose, i.e., gamers could play resource-intensive games. It’s worth noting that such dual-GPU setups faded away with the rapid development of newer GPUs. In short, single GPUs became powerful enough to overshadow dual GPUs. Eventually, a lot of applications and games stopped supporting CrossFire setups. With the new Intel Arc Pro B60, dual GPUs might make a mini-comeback. They’re not returning to the gaming arena, but to data centers, workstations, and AI setups. NVIDIA hasn’t promoted this in a long time, pushing its expensive GPUs into the market instead. This is also precisely where Intel wants to punch harder – the price point. Although we’re still waiting for the exact details on the Intel Arc Pro B60 dual GPU setup price, experts believe it could be priced around -To get the same 48 GB of VRAM, you might have to spend -6,000 with models like the NVIDIA RTX 6000 Ada. Of course, there will be performance differences between the two, too, but getting at-par VRAM is the first step of the process. Honestly, if power-intensive workspaces are able to get 48 GB of VRAM at 70-80% less cost, this dual GPU setup can become a market killer. In other news, Intel has also announced Project Battlematrix – a solution that supports AI workloads that can combine up to eight Intel Arc Pro GPUs in a system, with 192 GB of VRAM. The company seems to be working hard on the workstation and data processing industry – something NVIDIA and AMD haven’t catered to explicitly yet. This could give Intel a first-mover advantage. If it can achieve comparable performance with some modern NVIDIA chips, the new Intel Arc Pro B60 can revive the Blue team. : Nvidia’s downgraded H20 chips might not be enough to stop China’s Ai ambitions Krishi is a seasoned tech journalist with over four years of experience writing about PC hardware, consumer technology, and artificial intelligence.  Clarity and accessibility are at the core of Krishi’s writing style. He believes technology writing should empower readers—not confuse them—and he’s committed to ensuring his content is always easy to understand without sacrificing accuracy or depth. Over the years, Krishi has contributed to some of the most reputable names in the industry, including Techopedia, TechRadar, and Tom’s Guide. A man of many talents, Krishi has also proven his mettle as a crypto writer, tackling complex topics with both ease and zeal. His work spans various formats—from in-depth explainers and news coverage to feature pieces and buying guides.  Behind the scenes, Krishi operates from a dual-monitor setupthat’s always buzzing with news feeds, technical documentation, and research notes, as well as the occasional gaming sessions that keep him fresh.  Krishi thrives on staying current, always ready to dive into the latest announcements, industry shifts, and their far-reaching impacts.  When he's not deep into research on the latest PC hardware news, Krishi would love to chat with you about day trading and the financial markets—oh! And cricket, as well. View all articles by Krishi Chowdhary Our editorial process The Tech Report editorial policy is centered on providing helpful, accurate content that offers real value to our readers. We only work with experienced writers who have specific knowledge in the topics they cover, including latest developments in technology, online privacy, cryptocurrencies, software, and more. Our editorial policy ensures that each topic is researched and curated by our in-house editors. We maintain rigorous journalistic standards, and every article is 100% written by real authors. #intels #arc #pro #b60 #dual
    TECHREPORT.COM
    Is Intel’s Arc Pro B60 the Dual GPU Innovation We’ve Been Waiting For?
    Key Takeaways Maxsun unveiled the new Intel Arc Pro B60 Dual GPU video card, featuring two 24GB Pro B60 GPUs. It features a blower-style fan, ideal for building complex workstations with multiple cards. Although the pricing isn’t made public yet, we expect it to be around $1,000, which can be a competition-killer. Remember the good old CrossFire days when you could hook up multiple AMD GPUs to achieve higher performance? Well, those days might be returning thanks to the new Intel Arc Pro B60 Dual GPU launched by Maxsun at Computex. It PCIe features two 24GB GPUs in a single card, meaning you can get 48 GB of GPU memory. What’s incredible here is that the Arc Pro B60 only needs 8 PCIe lanes, unlike some high-end GPUs that need 16. However, you’ll need an X16 motherboard/CPU combo that supports PCIe bifurcation. In that case, the entire x16 slot can be split into two x8 connections. Interestingly, the Arc Pro B60 isn’t marketed as a gaming GPU – it’s meant primarily for workstations that require high computing power for tasks such as rendering, animations, moderate AI development, 3D modelling, etc. This is why it features a unique blower-like fan instead of the typical open-air fan we see in gaming CPUs. Unlike open-air fans that spread the heat out inside the case, a blower-style fan works like a hairdryer by pushing the heat out from the inside of the case. While the cooling efficiency of a blower fan might be lower on a single GPU, they turn out to be more effective in stacks, which is exactly what the Arc Pro B60 is meant for. So, if you’re building a workstation with 3 Arc Pro B60s, a blower-style fan ensures all 6 GPUs can be cooled without overwhelming the entire setup, making it better for ‘stacking.’ Comeback of Multi-GPU Setups Multi-GPU setups were quite the thing back in the day. We’re sure that a couple of die-hard builders in our team would serve up puppy eyes thinking about hooking up two 16GB GPUs to get more power. However, despite the widespread belief that you would get double the performance, memory was only enhanced by 30-50%. Still, these CrossFire setups served the purpose, i.e., gamers could play resource-intensive games. It’s worth noting that such dual-GPU setups faded away with the rapid development of newer GPUs. In short, single GPUs became powerful enough to overshadow dual GPUs. Eventually, a lot of applications and games stopped supporting CrossFire setups. With the new Intel Arc Pro B60, dual GPUs might make a mini-comeback. They’re not returning to the gaming arena, but to data centers, workstations, and AI setups. NVIDIA hasn’t promoted this in a long time, pushing its expensive GPUs into the market instead. This is also precisely where Intel wants to punch harder – the price point. Although we’re still waiting for the exact details on the Intel Arc Pro B60 dual GPU setup price, experts believe it could be priced around $1,000-$1,200. To get the same 48 GB of VRAM, you might have to spend $5,000-6,000 with models like the NVIDIA RTX 6000 Ada. Of course, there will be performance differences between the two, too, but getting at-par VRAM is the first step of the process. Honestly, if power-intensive workspaces are able to get 48 GB of VRAM at 70-80% less cost, this dual GPU setup can become a market killer. In other news, Intel has also announced Project Battlematrix – a solution that supports AI workloads that can combine up to eight Intel Arc Pro GPUs in a system, with 192 GB of VRAM. The company seems to be working hard on the workstation and data processing industry – something NVIDIA and AMD haven’t catered to explicitly yet. This could give Intel a first-mover advantage. If it can achieve comparable performance with some modern NVIDIA chips, the new Intel Arc Pro B60 can revive the Blue team. Read more: Nvidia’s downgraded H20 chips might not be enough to stop China’s Ai ambitions Krishi is a seasoned tech journalist with over four years of experience writing about PC hardware, consumer technology, and artificial intelligence.  Clarity and accessibility are at the core of Krishi’s writing style. He believes technology writing should empower readers—not confuse them—and he’s committed to ensuring his content is always easy to understand without sacrificing accuracy or depth. Over the years, Krishi has contributed to some of the most reputable names in the industry, including Techopedia, TechRadar, and Tom’s Guide. A man of many talents, Krishi has also proven his mettle as a crypto writer, tackling complex topics with both ease and zeal. His work spans various formats—from in-depth explainers and news coverage to feature pieces and buying guides.  Behind the scenes, Krishi operates from a dual-monitor setup (including a 29-inch LG UltraWide) that’s always buzzing with news feeds, technical documentation, and research notes, as well as the occasional gaming sessions that keep him fresh.  Krishi thrives on staying current, always ready to dive into the latest announcements, industry shifts, and their far-reaching impacts.  When he's not deep into research on the latest PC hardware news, Krishi would love to chat with you about day trading and the financial markets—oh! And cricket, as well. View all articles by Krishi Chowdhary Our editorial process The Tech Report editorial policy is centered on providing helpful, accurate content that offers real value to our readers. We only work with experienced writers who have specific knowledge in the topics they cover, including latest developments in technology, online privacy, cryptocurrencies, software, and more. Our editorial policy ensures that each topic is researched and curated by our in-house editors. We maintain rigorous journalistic standards, and every article is 100% written by real authors.
    0 Comentários 0 Compartilhamentos 0 Anterior
  • Intel’s Next Battlemage Gaming GPU, the Arc B770, Is Still Expected To Release In Q4 2025; Team Blue Won’t Give Up Anytime Soon

    Intel still has plans for the desktop GPU segment, as it is reported that the company's next Arc B770 GPU is slated to launch by the fourth quarter of this year.
    Intel Still Has Plans To Make An Impact In The Gaming GPU Segment, Plans To Unveil Arc B770 In Upcoming Months
    At Computex, we were hoping for some kind of consumer-oriented release by Intel, since there were rumors that the firm could unveil a new Battlemage variant. Team Blue unveiled GPUs, but they weren't for gaming. Instead, the company expanded its focus towards the professional segment, adding new models to its Arc Pro lineup. However, according to a report by Tweakers, Intel hasn't abandoned the gaming segment at all, and instead plans to unveil the Arc B770 in the upcoming quarters.
    Intel is working on an Arc B770 video card based on the Battlemage architecture. Tweakers was able to confirm this with multiple sources during Computex.
    Sources close to Intel confirm the existence of the Arc B770 video card. According to them, the introduction is planned for the second half of this year. The card is expected to appear in the fourth quarter, although the planning can change.

    - Tweakers

    Intel hasn't been very vocal about its plans for the dGPU segment, since the last release we saw from the company was the Arc B580 and the Arc B570, which were two Battlemage variants focused on the mid-tier segment. However, Intel has been busy developing a solution around the BMG-G31 silicon, which is expected to power the Arc B770, so it was safe to assume that we would see an announcement at Computex. But this has been pushed ahead, and we might have a showcase at Intel Innovation 2025, which is expected to be held by September.

    As to what we can expect from it, a variant on this die was said to feature around 24-32 Xe2 cores with a 256-bit memory bus and 16 GB of GDDR6 memory, making it a formidable rival to NVIDIA's 60-class series and AMD's RX 9060 lineup. Considering Intel's track record in the gaming GPU market, the firm should focus on making an impact, especially on the mainstream GPU segment, since the competition is ramping up.

    Deal of the Day
    #intels #next #battlemage #gaming #gpu
    Intel’s Next Battlemage Gaming GPU, the Arc B770, Is Still Expected To Release In Q4 2025; Team Blue Won’t Give Up Anytime Soon
    Intel still has plans for the desktop GPU segment, as it is reported that the company's next Arc B770 GPU is slated to launch by the fourth quarter of this year. Intel Still Has Plans To Make An Impact In The Gaming GPU Segment, Plans To Unveil Arc B770 In Upcoming Months At Computex, we were hoping for some kind of consumer-oriented release by Intel, since there were rumors that the firm could unveil a new Battlemage variant. Team Blue unveiled GPUs, but they weren't for gaming. Instead, the company expanded its focus towards the professional segment, adding new models to its Arc Pro lineup. However, according to a report by Tweakers, Intel hasn't abandoned the gaming segment at all, and instead plans to unveil the Arc B770 in the upcoming quarters. Intel is working on an Arc B770 video card based on the Battlemage architecture. Tweakers was able to confirm this with multiple sources during Computex. Sources close to Intel confirm the existence of the Arc B770 video card. According to them, the introduction is planned for the second half of this year. The card is expected to appear in the fourth quarter, although the planning can change. - Tweakers Intel hasn't been very vocal about its plans for the dGPU segment, since the last release we saw from the company was the Arc B580 and the Arc B570, which were two Battlemage variants focused on the mid-tier segment. However, Intel has been busy developing a solution around the BMG-G31 silicon, which is expected to power the Arc B770, so it was safe to assume that we would see an announcement at Computex. But this has been pushed ahead, and we might have a showcase at Intel Innovation 2025, which is expected to be held by September. As to what we can expect from it, a variant on this die was said to feature around 24-32 Xe2 cores with a 256-bit memory bus and 16 GB of GDDR6 memory, making it a formidable rival to NVIDIA's 60-class series and AMD's RX 9060 lineup. Considering Intel's track record in the gaming GPU market, the firm should focus on making an impact, especially on the mainstream GPU segment, since the competition is ramping up. Deal of the Day #intels #next #battlemage #gaming #gpu
    WCCFTECH.COM
    Intel’s Next Battlemage Gaming GPU, the Arc B770, Is Still Expected To Release In Q4 2025; Team Blue Won’t Give Up Anytime Soon
    Intel still has plans for the desktop GPU segment, as it is reported that the company's next Arc B770 GPU is slated to launch by the fourth quarter of this year. Intel Still Has Plans To Make An Impact In The Gaming GPU Segment, Plans To Unveil Arc B770 In Upcoming Months At Computex, we were hoping for some kind of consumer-oriented release by Intel, since there were rumors that the firm could unveil a new Battlemage variant. Team Blue unveiled GPUs, but they weren't for gaming. Instead, the company expanded its focus towards the professional segment, adding new models to its Arc Pro lineup. However, according to a report by Tweakers, Intel hasn't abandoned the gaming segment at all, and instead plans to unveil the Arc B770 in the upcoming quarters. Intel is working on an Arc B770 video card based on the Battlemage architecture. Tweakers was able to confirm this with multiple sources during Computex. Sources close to Intel confirm the existence of the Arc B770 video card. According to them, the introduction is planned for the second half of this year. The card is expected to appear in the fourth quarter, although the planning can change. - Tweakers Intel hasn't been very vocal about its plans for the dGPU segment, since the last release we saw from the company was the Arc B580 and the Arc B570, which were two Battlemage variants focused on the mid-tier segment. However, Intel has been busy developing a solution around the BMG-G31 silicon, which is expected to power the Arc B770, so it was safe to assume that we would see an announcement at Computex. But this has been pushed ahead, and we might have a showcase at Intel Innovation 2025, which is expected to be held by September. As to what we can expect from it, a variant on this die was said to feature around 24-32 Xe2 cores with a 256-bit memory bus and 16 GB of GDDR6 memory, making it a formidable rival to NVIDIA's 60-class series and AMD's RX 9060 lineup. Considering Intel's track record in the gaming GPU market, the firm should focus on making an impact, especially on the mainstream GPU segment, since the competition is ramping up. Deal of the Day
    0 Comentários 0 Compartilhamentos 0 Anterior
  • Starfish Space Announces Plans For First Commercial Satellite Docking

    Starfish Space plans to perform the first commercial satellite docking in orbit with its Otter Pup 2 mission, aiming to connect to an unprepared D-Orbit ION spacecraft using an electrostatic capture mechanism and autonomous navigation software. NASASpaceFlight.com reports: This follows the company's first attempt, which saw the Otter Pup 1 mission unable to dock with its target due to a thruster failure. The Otter Pup 2 spacecraft will be deployed from a quarter plate on the upper stage adapter of the SpaceX Falcon 9 rocket, placing it into a sun synchronous orbit altitude of 510 km inclined 97.4 degrees. The target will be a D-Orbit ION spacecraft which will simulate a client payload, which is not equipped with a traditional docking adapter or capture plate as you might see aboard a space station or other rendezvous target. Instead, Starfish Space's Nautilus capture mechanism will feature a special end effector connected to the end of the capture mechanism. This end effector will enable Otter Pup 2 to dock with the ION through electrostatic adhesion.

    "An electromagnet will be integrated into the end effector and will be used as a backup option to the electrostatic end effector, to dock with the ION through magnetic attraction," the company notes. The goal is to eventually commission its Otter satellite servicing vehicle to allow for servicing of previously launched satellites. The company's first Otter missions include customers such as NASA, the U.S. Space Force, and Intelsat, with the goal of flying those missions as soon as 2026.Following the thruster issues on the first mission, this flight will feature two ThrustMe thrusters, which use an electric propulsion system based on gridded ion thruster technology.

    of this story at Slashdot.
    #starfish #space #announces #plans #first
    Starfish Space Announces Plans For First Commercial Satellite Docking
    Starfish Space plans to perform the first commercial satellite docking in orbit with its Otter Pup 2 mission, aiming to connect to an unprepared D-Orbit ION spacecraft using an electrostatic capture mechanism and autonomous navigation software. NASASpaceFlight.com reports: This follows the company's first attempt, which saw the Otter Pup 1 mission unable to dock with its target due to a thruster failure. The Otter Pup 2 spacecraft will be deployed from a quarter plate on the upper stage adapter of the SpaceX Falcon 9 rocket, placing it into a sun synchronous orbit altitude of 510 km inclined 97.4 degrees. The target will be a D-Orbit ION spacecraft which will simulate a client payload, which is not equipped with a traditional docking adapter or capture plate as you might see aboard a space station or other rendezvous target. Instead, Starfish Space's Nautilus capture mechanism will feature a special end effector connected to the end of the capture mechanism. This end effector will enable Otter Pup 2 to dock with the ION through electrostatic adhesion. "An electromagnet will be integrated into the end effector and will be used as a backup option to the electrostatic end effector, to dock with the ION through magnetic attraction," the company notes. The goal is to eventually commission its Otter satellite servicing vehicle to allow for servicing of previously launched satellites. The company's first Otter missions include customers such as NASA, the U.S. Space Force, and Intelsat, with the goal of flying those missions as soon as 2026.Following the thruster issues on the first mission, this flight will feature two ThrustMe thrusters, which use an electric propulsion system based on gridded ion thruster technology. of this story at Slashdot. #starfish #space #announces #plans #first
    SCIENCE.SLASHDOT.ORG
    Starfish Space Announces Plans For First Commercial Satellite Docking
    Starfish Space plans to perform the first commercial satellite docking in orbit with its Otter Pup 2 mission, aiming to connect to an unprepared D-Orbit ION spacecraft using an electrostatic capture mechanism and autonomous navigation software. NASASpaceFlight.com reports: This follows the company's first attempt, which saw the Otter Pup 1 mission unable to dock with its target due to a thruster failure. The Otter Pup 2 spacecraft will be deployed from a quarter plate on the upper stage adapter of the SpaceX Falcon 9 rocket, placing it into a sun synchronous orbit altitude of 510 km inclined 97.4 degrees. The target will be a D-Orbit ION spacecraft which will simulate a client payload, which is not equipped with a traditional docking adapter or capture plate as you might see aboard a space station or other rendezvous target. Instead, Starfish Space's Nautilus capture mechanism will feature a special end effector connected to the end of the capture mechanism. This end effector will enable Otter Pup 2 to dock with the ION through electrostatic adhesion. "An electromagnet will be integrated into the end effector and will be used as a backup option to the electrostatic end effector, to dock with the ION through magnetic attraction," the company notes. The goal is to eventually commission its Otter satellite servicing vehicle to allow for servicing of previously launched satellites. The company's first Otter missions include customers such as NASA, the U.S. Space Force, and Intelsat, with the goal of flying those missions as soon as 2026. [...] Following the thruster issues on the first mission, this flight will feature two ThrustMe thrusters, which use an electric propulsion system based on gridded ion thruster technology. Read more of this story at Slashdot.
    1 Comentários 0 Compartilhamentos 0 Anterior
  • Agentic AI 102: Guardrails and Agent Evaluation

    Introduction

    In the first post of this series, we talked about the fundamentals of creating AI Agents and introduced concepts like reasoning, memory, and tools.

    Of course, that first post touched only the surface of this new area of the data industry. There is so much more that can be done, and we are going to learn more along the way in this series.

    So, it is time to take one step further.

    In this post, we will cover three topics:

    Guardrails: these are safe blocks that prevent a Large Language Modelfrom responding about some topics.

    Agent Evaluation: Have you ever thought about how accurate the responses from LLM are? I bet you did. So we will see the main ways to measure that.

    Monitoring: We will also learn about the built-in monitoring app in Agno’s framework.

    We shall begin now.

    Guardrails

    Our first topic is the simplest, in my opinion. Guardrails are rules that will keep an AI agent from responding to a given topic or list of topics.

    I believe there is a good chance that you have ever asked something to ChatGPT or Gemini and received a response like “I can’t talk about this topic”, or “Please consult a professional specialist”, something like that. Usually, that occurs with sensitive topics like health advice, psychological conditions, or financial advice.

    Those blocks are safeguards to prevent people from hurting themselves, harming their health, or their pockets. As we know, LLMs are trained on massive amounts of text, ergo inheriting a lot of bad content with it, which could easily lead to bad advice in those areas for people. And I didn’t even mention hallucinations!

    Think about how many stories there are of people who lost money by following investment tips from online forums. Or how many people took the wrong medicine because they read about it on the internet.

    Well, I guess you got the point. We must prevent our agents from talking about certain topics or taking certain actions. For that, we will use guardrails.

    The best framework I found to impose those blocks is Guardrails AI. There, you will see a hub full of predefined rules that a response must follow in order to pass and be displayed to the user.

    To get started quickly, first go to this linkand get an API key. Then, install the package. Next, type the guardrails setup command. It will ask you a couple of questions that you can respond n, and it will ask you to enter the API Key generated.

    pip install guardrails-ai
    guardrails configure

    Once that is completed, go to the Guardrails AI Huband choose one that you need. Every guardrail has instructions on how to implement it. Basically, you install it via the command line and then use it like a module in Python.

    For this example, we’re choosing one called Restrict to Topic, which, as its name says, lets the user talk only about what’s in the list. So, go back to the terminal and install it using the code below.

    guardrails hub install hub://tryolabs/restricttotopic

    Next, let’s open our Python script and import some modules.

    # Imports
    from agno.agent import Agent
    from agno.models.google import Gemini
    import os

    # Import Guard and Validator
    from guardrails import Guard
    from guardrails.hub import RestrictToTopic

    Next, we create the guard. We will restrict our agent to talk only about sports or the weather. And we are restricting it to talk about stocks.

    # Setup Guard
    guard = Guard.use)

    Now we can run the agent and the guard.

    # Create agent
    agent = Agent),
    description= "An assistant agent",
    instructions=,
    markdown= True
    )

    # Run the agent
    response = agent.run.content

    # Run agent with validation
    validation_step = guard.validate# Print validated response
    if validation_step.validation_passed:
    printelse:
    printThis is the response when we ask about a stock symbol.

    Validation Failed Invalid topics found:If I ask about a topic that is not on the valid_topics list, I will also see a block.

    "What's the number one soda drink?"
    Validation Failed No valid topic was found.

    Finally, let’s ask about sports.

    "Who is Michael Jordan?"
    Michael Jordan is a former professional basketball player widely considered one of
    the greatest of all time. He won six NBA championships with the Chicago Bulls.

    And we saw a response this time, as it is a valid topic.

    Let’s move on to the evaluation of agents now.

    Agent Evaluation

    Since I started studying LLMs and Agentic Ai, one of my main questions has been about model evaluation. Unlike traditional Data Science Modeling, where you have structured metrics that are adequate for each case, for AI Agents, this is more blurry.

    Fortunately, the developer community is pretty quick in finding solutions for almost everything, and so they created this nice package for LLMs evaluation: deepeval.

    DeepEvalis a library created by Confident AI that gathers many methods to evaluate LLMs and AI Agents. In this section, let’s learn a couple of the main methods, just so we can build some intuition on the subject, and also because the library is quite extensive.\

    The first evaluation is the most basic we can use, and it is called G-Eval. As AI tools like ChatGPT become more common in everyday tasks, we have to make sure they’re giving helpful and accurate responses. That’s where G-Eval from the DeepEval Python package comes in.

    G-Eval is like a smart reviewer that uses another AI model to evaluate how well a chatbot or AI assistant is performing. For example. My agent runs Gemini, and I am using OpenAI to assess it. This method takes a more advanced approach than a human one by asking an AI to “grade” another AI’s answers based on things like relevance, correctness, and clarity.

    It’s a nice way to test and improve generative AI systems in a more scalable way. Let’s quickly code an example. We will import the modules, create a prompt, a simple chat agent, and ask it about a description of the weather for the month of May in NYC.

    # Imports
    from agno.agent import Agent
    from agno.models.google import Gemini
    import os
    # Evaluation Modules
    from deepeval.test_case import LLMTestCase, LLMTestCaseParams
    from deepeval.metrics import GEval

    # Prompt
    prompt = "Describe the weather in NYC for May"

    # Create agent
    agent = Agent),
    description= "An assistant agent",
    instructions=,
    markdown= True,
    monitoring= True
    )

    # Run agent
    response = agent.run# Print response
    printIt responds: “Mild, with average highs in the 60s°F and lows in the 50s°F. Expect some rain“.

    Nice. Seems pretty good to me.

    But how can we put a number on it and show a potential manager or client how our agent is doing?

    Here is how:

    Create a test case passing the prompt and the response to the LLMTestCase class.

    Create a metric. We will use the method GEval and add a prompt for the model to test it for coherence, and then I give it the meaning of what coherence is to me.

    Give the output as evaluation_params.

    Run the measure method and get the score and reason from it.

    # Test Case
    test_case = LLMTestCase# Setup the Metric
    coherence_metric = GEval# Run the metric
    coherence_metric.measureprintprintThe output looks like this.

    0.9
    The response directly addresses the prompt about NYC weather in May,
    maintains logical consistency, flows naturally, and uses clear language.
    However, it could be slightly more detailed.

    0.9 seems pretty good, given that the default threshold is 0.5.

    If you want to check the logs, use this next snippet.

    # Check the logs
    printHere’s the response.

    Criteria:
    Coherence. The agent can answer the prompt and the response makes sense.

    Evaluation Steps:Very nice. Now let us learn about another interesting use case, which is the evaluation of task completion for AI Agents. Elaborating a little more, how our agent is doing when it is requested to perform a task, and how much of it the agent can deliver.

    First, we are creating a simple agent that can access Wikipedia and summarize the topic of the query.

    # Imports
    from agno.agent import Agent
    from agno.models.google import Gemini
    from agno.tools.wikipedia import WikipediaTools
    import os
    from deepeval.test_case import LLMTestCase, ToolCall
    from deepeval.metrics import TaskCompletionMetric
    from deepeval import evaluate

    # Prompt
    prompt = "Search wikipedia for 'Time series analysis' and summarize the 3 main points"

    # Create agent
    agent = Agent),
    description= "You are a researcher specialized in searching the wikipedia.",
    tools=,
    show_tool_calls= True,
    markdown= True,
    read_tool_call_history= True
    )

    # Run agent
    response = agent.run# Print response
    printThe result looks very good. Let’s evaluate it using the TaskCompletionMetric class.

    # Create a Metric
    metric = TaskCompletionMetric# Test Case
    test_case = LLMTestCase]
    )

    # Evaluate
    evaluateOutput, including the agent’s response.

    ======================================================================

    Metrics Summary

    - Task CompletionFor test case:

    - input: Search wikipedia for 'Time series analysis' and summarize the 3 main points
    - actual output: Here are the 3 main points about Time series analysis based on the
    Wikipedia search:

    1. **Definition:** A time series is a sequence of data points indexed in time order,
    often taken at successive, equally spaced points in time.
    2. **Applications:** Time series analysis is used in various fields like statistics,
    signal processing, econometrics, weather forecasting, and more, wherever temporal
    measurements are involved.
    3. **Purpose:** Time series analysis involves methods for extracting meaningful
    statistics and characteristics from time series data, and time series forecasting
    uses models to predict future values based on past observations.

    - expected output: None
    - context: None
    - retrieval context: None

    ======================================================================

    Overall Metric Pass Rates

    Task Completion: 100.00% pass rate

    ======================================================================

    ✓ Tests finished ! Run 'deepeval login' to save and analyze evaluation results
    on Confident AI.

    Our agent passed the test with honor: 100%!

    You can learn much more about the DeepEval library in this link.

    Finally, in the next section, we will learn the capabilities of Agno’s library for monitoring agents.

    Agent Monitoring

    Like I told you in my previous post, I chose Agno to learn more about Agentic AI. Just to be clear, this is not a sponsored post. It is just that I think this is the best option for those starting their journey learning about this topic.

    So, one of the cool things we can take advantage of using Agno’s framework is the app they make available for model monitoring.

    Take this agent that can search the internet and write Instagram posts, for example.

    # Imports
    import os
    from agno.agent import Agent
    from agno.models.google import Gemini
    from agno.tools.file import FileTools
    from agno.tools.googlesearch import GoogleSearchTools

    # Topic
    topic = "Healthy Eating"

    # Create agent
    agent = Agent),
    description= f"""You are a social media marketer specialized in creating engaging content.
    Search the internet for 'trending topics about {topic}' and use them to create a post.""",
    tools=,
    expected_output="""A short post for instagram and a prompt for a picture related to the content of the post.
    Don't use emojis or special characters in the post. If you find an error in the character encoding, remove the character before saving the file.
    Use the template:
    - Post
    - Prompt for the picture
    the post to a file named 'post.txt'.""",
    show_tool_calls=True,
    monitoring=True)

    # Writing and saving a file
    agent.print_responseTo monitor its performance, follow these steps:

    Go to and get an API Key.

    Open a terminal and type ag setup.

    If it is the first time, it might ask for the API Key. Copy and Paste it in the terminal prompt.

    You will see the Dashboard tab open in your browser.

    If you want to monitor your agent, add the argument monitoring=True.

    Run your agent.

    Go to the Dashboard on the web browser.

    Click on Sessions. As it is a single agent, you will see it under the tab Agents on the top portion of the page.

    Agno Dashboard after running the agent. Image by the author.

    The cools features we can see there are:

    Info about the model

    The response

    Tools used

    Tokens consumed

    This is the resulting token consumption while saving the file. Image by the author.

    Pretty neat, huh?

    This is useful for us to know where the agent is spending more or less tokens, and where it is taking more time to perform a task, for example.

    Well, let’s wrap up then.

    Before You Go

    We have learned a lot in this second round. In this post, we covered:

    Guardrails for AI are essential safety measures and ethical guidelines implemented to prevent unintended harmful outputs and ensure responsible AI behavior.

    Model evaluation, exemplified by GEval for broad assessment and TaskCompletion with DeepEval for agents output quality, is crucial for understanding AI capabilities and limitations.

    Model monitoring with Agno’s app, including tracking token usage and response time, which is vital for managing costs, ensuring performance, and identifying potential issues in deployed AI systems.

    Contact & Follow Me

    If you liked this content, find more of my work in my website.



    GitHub Repository



    References//

    The post Agentic AI 102: Guardrails and Agent Evaluation appeared first on Towards Data Science.
    #agentic #guardrails #agent #evaluation
    Agentic AI 102: Guardrails and Agent Evaluation
    Introduction In the first post of this series, we talked about the fundamentals of creating AI Agents and introduced concepts like reasoning, memory, and tools. Of course, that first post touched only the surface of this new area of the data industry. There is so much more that can be done, and we are going to learn more along the way in this series. So, it is time to take one step further. In this post, we will cover three topics: Guardrails: these are safe blocks that prevent a Large Language Modelfrom responding about some topics. Agent Evaluation: Have you ever thought about how accurate the responses from LLM are? I bet you did. So we will see the main ways to measure that. Monitoring: We will also learn about the built-in monitoring app in Agno’s framework. We shall begin now. Guardrails Our first topic is the simplest, in my opinion. Guardrails are rules that will keep an AI agent from responding to a given topic or list of topics. I believe there is a good chance that you have ever asked something to ChatGPT or Gemini and received a response like “I can’t talk about this topic”, or “Please consult a professional specialist”, something like that. Usually, that occurs with sensitive topics like health advice, psychological conditions, or financial advice. Those blocks are safeguards to prevent people from hurting themselves, harming their health, or their pockets. As we know, LLMs are trained on massive amounts of text, ergo inheriting a lot of bad content with it, which could easily lead to bad advice in those areas for people. And I didn’t even mention hallucinations! Think about how many stories there are of people who lost money by following investment tips from online forums. Or how many people took the wrong medicine because they read about it on the internet. Well, I guess you got the point. We must prevent our agents from talking about certain topics or taking certain actions. For that, we will use guardrails. The best framework I found to impose those blocks is Guardrails AI. There, you will see a hub full of predefined rules that a response must follow in order to pass and be displayed to the user. To get started quickly, first go to this linkand get an API key. Then, install the package. Next, type the guardrails setup command. It will ask you a couple of questions that you can respond n, and it will ask you to enter the API Key generated. pip install guardrails-ai guardrails configure Once that is completed, go to the Guardrails AI Huband choose one that you need. Every guardrail has instructions on how to implement it. Basically, you install it via the command line and then use it like a module in Python. For this example, we’re choosing one called Restrict to Topic, which, as its name says, lets the user talk only about what’s in the list. So, go back to the terminal and install it using the code below. guardrails hub install hub://tryolabs/restricttotopic Next, let’s open our Python script and import some modules. # Imports from agno.agent import Agent from agno.models.google import Gemini import os # Import Guard and Validator from guardrails import Guard from guardrails.hub import RestrictToTopic Next, we create the guard. We will restrict our agent to talk only about sports or the weather. And we are restricting it to talk about stocks. # Setup Guard guard = Guard.use) Now we can run the agent and the guard. # Create agent agent = Agent), description= "An assistant agent", instructions=, markdown= True ) # Run the agent response = agent.run.content # Run agent with validation validation_step = guard.validate# Print validated response if validation_step.validation_passed: printelse: printThis is the response when we ask about a stock symbol. Validation Failed Invalid topics found:If I ask about a topic that is not on the valid_topics list, I will also see a block. "What's the number one soda drink?" Validation Failed No valid topic was found. Finally, let’s ask about sports. "Who is Michael Jordan?" Michael Jordan is a former professional basketball player widely considered one of the greatest of all time. He won six NBA championships with the Chicago Bulls. And we saw a response this time, as it is a valid topic. Let’s move on to the evaluation of agents now. Agent Evaluation Since I started studying LLMs and Agentic Ai, one of my main questions has been about model evaluation. Unlike traditional Data Science Modeling, where you have structured metrics that are adequate for each case, for AI Agents, this is more blurry. Fortunately, the developer community is pretty quick in finding solutions for almost everything, and so they created this nice package for LLMs evaluation: deepeval. DeepEvalis a library created by Confident AI that gathers many methods to evaluate LLMs and AI Agents. In this section, let’s learn a couple of the main methods, just so we can build some intuition on the subject, and also because the library is quite extensive.\ The first evaluation is the most basic we can use, and it is called G-Eval. As AI tools like ChatGPT become more common in everyday tasks, we have to make sure they’re giving helpful and accurate responses. That’s where G-Eval from the DeepEval Python package comes in. G-Eval is like a smart reviewer that uses another AI model to evaluate how well a chatbot or AI assistant is performing. For example. My agent runs Gemini, and I am using OpenAI to assess it. This method takes a more advanced approach than a human one by asking an AI to “grade” another AI’s answers based on things like relevance, correctness, and clarity. It’s a nice way to test and improve generative AI systems in a more scalable way. Let’s quickly code an example. We will import the modules, create a prompt, a simple chat agent, and ask it about a description of the weather for the month of May in NYC. # Imports from agno.agent import Agent from agno.models.google import Gemini import os # Evaluation Modules from deepeval.test_case import LLMTestCase, LLMTestCaseParams from deepeval.metrics import GEval # Prompt prompt = "Describe the weather in NYC for May" # Create agent agent = Agent), description= "An assistant agent", instructions=, markdown= True, monitoring= True ) # Run agent response = agent.run# Print response printIt responds: “Mild, with average highs in the 60s°F and lows in the 50s°F. Expect some rain“. Nice. Seems pretty good to me. But how can we put a number on it and show a potential manager or client how our agent is doing? Here is how: Create a test case passing the prompt and the response to the LLMTestCase class. Create a metric. We will use the method GEval and add a prompt for the model to test it for coherence, and then I give it the meaning of what coherence is to me. Give the output as evaluation_params. Run the measure method and get the score and reason from it. # Test Case test_case = LLMTestCase# Setup the Metric coherence_metric = GEval# Run the metric coherence_metric.measureprintprintThe output looks like this. 0.9 The response directly addresses the prompt about NYC weather in May, maintains logical consistency, flows naturally, and uses clear language. However, it could be slightly more detailed. 0.9 seems pretty good, given that the default threshold is 0.5. If you want to check the logs, use this next snippet. # Check the logs printHere’s the response. Criteria: Coherence. The agent can answer the prompt and the response makes sense. Evaluation Steps:Very nice. Now let us learn about another interesting use case, which is the evaluation of task completion for AI Agents. Elaborating a little more, how our agent is doing when it is requested to perform a task, and how much of it the agent can deliver. First, we are creating a simple agent that can access Wikipedia and summarize the topic of the query. # Imports from agno.agent import Agent from agno.models.google import Gemini from agno.tools.wikipedia import WikipediaTools import os from deepeval.test_case import LLMTestCase, ToolCall from deepeval.metrics import TaskCompletionMetric from deepeval import evaluate # Prompt prompt = "Search wikipedia for 'Time series analysis' and summarize the 3 main points" # Create agent agent = Agent), description= "You are a researcher specialized in searching the wikipedia.", tools=, show_tool_calls= True, markdown= True, read_tool_call_history= True ) # Run agent response = agent.run# Print response printThe result looks very good. Let’s evaluate it using the TaskCompletionMetric class. # Create a Metric metric = TaskCompletionMetric# Test Case test_case = LLMTestCase] ) # Evaluate evaluateOutput, including the agent’s response. ====================================================================== Metrics Summary - Task CompletionFor test case: - input: Search wikipedia for 'Time series analysis' and summarize the 3 main points - actual output: Here are the 3 main points about Time series analysis based on the Wikipedia search: 1. **Definition:** A time series is a sequence of data points indexed in time order, often taken at successive, equally spaced points in time. 2. **Applications:** Time series analysis is used in various fields like statistics, signal processing, econometrics, weather forecasting, and more, wherever temporal measurements are involved. 3. **Purpose:** Time series analysis involves methods for extracting meaningful statistics and characteristics from time series data, and time series forecasting uses models to predict future values based on past observations. - expected output: None - context: None - retrieval context: None ====================================================================== Overall Metric Pass Rates Task Completion: 100.00% pass rate ====================================================================== ✓ Tests finished ! Run 'deepeval login' to save and analyze evaluation results on Confident AI. Our agent passed the test with honor: 100%! You can learn much more about the DeepEval library in this link. Finally, in the next section, we will learn the capabilities of Agno’s library for monitoring agents. Agent Monitoring Like I told you in my previous post, I chose Agno to learn more about Agentic AI. Just to be clear, this is not a sponsored post. It is just that I think this is the best option for those starting their journey learning about this topic. So, one of the cool things we can take advantage of using Agno’s framework is the app they make available for model monitoring. Take this agent that can search the internet and write Instagram posts, for example. # Imports import os from agno.agent import Agent from agno.models.google import Gemini from agno.tools.file import FileTools from agno.tools.googlesearch import GoogleSearchTools # Topic topic = "Healthy Eating" # Create agent agent = Agent), description= f"""You are a social media marketer specialized in creating engaging content. Search the internet for 'trending topics about {topic}' and use them to create a post.""", tools=, expected_output="""A short post for instagram and a prompt for a picture related to the content of the post. Don't use emojis or special characters in the post. If you find an error in the character encoding, remove the character before saving the file. Use the template: - Post - Prompt for the picture the post to a file named 'post.txt'.""", show_tool_calls=True, monitoring=True) # Writing and saving a file agent.print_responseTo monitor its performance, follow these steps: Go to and get an API Key. Open a terminal and type ag setup. If it is the first time, it might ask for the API Key. Copy and Paste it in the terminal prompt. You will see the Dashboard tab open in your browser. If you want to monitor your agent, add the argument monitoring=True. Run your agent. Go to the Dashboard on the web browser. Click on Sessions. As it is a single agent, you will see it under the tab Agents on the top portion of the page. Agno Dashboard after running the agent. Image by the author. The cools features we can see there are: Info about the model The response Tools used Tokens consumed This is the resulting token consumption while saving the file. Image by the author. Pretty neat, huh? This is useful for us to know where the agent is spending more or less tokens, and where it is taking more time to perform a task, for example. Well, let’s wrap up then. Before You Go We have learned a lot in this second round. In this post, we covered: Guardrails for AI are essential safety measures and ethical guidelines implemented to prevent unintended harmful outputs and ensure responsible AI behavior. Model evaluation, exemplified by GEval for broad assessment and TaskCompletion with DeepEval for agents output quality, is crucial for understanding AI capabilities and limitations. Model monitoring with Agno’s app, including tracking token usage and response time, which is vital for managing costs, ensuring performance, and identifying potential issues in deployed AI systems. Contact & Follow Me If you liked this content, find more of my work in my website. GitHub Repository References// The post Agentic AI 102: Guardrails and Agent Evaluation appeared first on Towards Data Science. #agentic #guardrails #agent #evaluation
    TOWARDSDATASCIENCE.COM
    Agentic AI 102: Guardrails and Agent Evaluation
    Introduction In the first post of this series (Agentic AI 101: Starting Your Journey Building AI Agents), we talked about the fundamentals of creating AI Agents and introduced concepts like reasoning, memory, and tools. Of course, that first post touched only the surface of this new area of the data industry. There is so much more that can be done, and we are going to learn more along the way in this series. So, it is time to take one step further. In this post, we will cover three topics: Guardrails: these are safe blocks that prevent a Large Language Model (LLM) from responding about some topics. Agent Evaluation: Have you ever thought about how accurate the responses from LLM are? I bet you did. So we will see the main ways to measure that. Monitoring: We will also learn about the built-in monitoring app in Agno’s framework. We shall begin now. Guardrails Our first topic is the simplest, in my opinion. Guardrails are rules that will keep an AI agent from responding to a given topic or list of topics. I believe there is a good chance that you have ever asked something to ChatGPT or Gemini and received a response like “I can’t talk about this topic”, or “Please consult a professional specialist”, something like that. Usually, that occurs with sensitive topics like health advice, psychological conditions, or financial advice. Those blocks are safeguards to prevent people from hurting themselves, harming their health, or their pockets. As we know, LLMs are trained on massive amounts of text, ergo inheriting a lot of bad content with it, which could easily lead to bad advice in those areas for people. And I didn’t even mention hallucinations! Think about how many stories there are of people who lost money by following investment tips from online forums. Or how many people took the wrong medicine because they read about it on the internet. Well, I guess you got the point. We must prevent our agents from talking about certain topics or taking certain actions. For that, we will use guardrails. The best framework I found to impose those blocks is Guardrails AI [1]. There, you will see a hub full of predefined rules that a response must follow in order to pass and be displayed to the user. To get started quickly, first go to this link [2] and get an API key. Then, install the package. Next, type the guardrails setup command. It will ask you a couple of questions that you can respond n (for No), and it will ask you to enter the API Key generated. pip install guardrails-ai guardrails configure Once that is completed, go to the Guardrails AI Hub [3] and choose one that you need. Every guardrail has instructions on how to implement it. Basically, you install it via the command line and then use it like a module in Python. For this example, we’re choosing one called Restrict to Topic [4], which, as its name says, lets the user talk only about what’s in the list. So, go back to the terminal and install it using the code below. guardrails hub install hub://tryolabs/restricttotopic Next, let’s open our Python script and import some modules. # Imports from agno.agent import Agent from agno.models.google import Gemini import os # Import Guard and Validator from guardrails import Guard from guardrails.hub import RestrictToTopic Next, we create the guard. We will restrict our agent to talk only about sports or the weather. And we are restricting it to talk about stocks. # Setup Guard guard = Guard().use( RestrictToTopic( valid_topics=["sports", "weather"], invalid_topics=["stocks"], disable_classifier=True, disable_llm=False, on_fail="filter" ) ) Now we can run the agent and the guard. # Create agent agent = Agent( model= Gemini(id="gemini-1.5-flash", api_key = os.environ.get("GEMINI_API_KEY")), description= "An assistant agent", instructions= ["Be sucint. Reply in maximum two sentences"], markdown= True ) # Run the agent response = agent.run("What's the ticker symbol for Apple?").content # Run agent with validation validation_step = guard.validate(response) # Print validated response if validation_step.validation_passed: print(response) else: print("Validation Failed", validation_step.validation_summaries[0].failure_reason) This is the response when we ask about a stock symbol. Validation Failed Invalid topics found: ['stocks'] If I ask about a topic that is not on the valid_topics list, I will also see a block. "What's the number one soda drink?" Validation Failed No valid topic was found. Finally, let’s ask about sports. "Who is Michael Jordan?" Michael Jordan is a former professional basketball player widely considered one of the greatest of all time. He won six NBA championships with the Chicago Bulls. And we saw a response this time, as it is a valid topic. Let’s move on to the evaluation of agents now. Agent Evaluation Since I started studying LLMs and Agentic Ai, one of my main questions has been about model evaluation. Unlike traditional Data Science Modeling, where you have structured metrics that are adequate for each case, for AI Agents, this is more blurry. Fortunately, the developer community is pretty quick in finding solutions for almost everything, and so they created this nice package for LLMs evaluation: deepeval. DeepEval [5] is a library created by Confident AI that gathers many methods to evaluate LLMs and AI Agents. In this section, let’s learn a couple of the main methods, just so we can build some intuition on the subject, and also because the library is quite extensive.\ The first evaluation is the most basic we can use, and it is called G-Eval. As AI tools like ChatGPT become more common in everyday tasks, we have to make sure they’re giving helpful and accurate responses. That’s where G-Eval from the DeepEval Python package comes in. G-Eval is like a smart reviewer that uses another AI model to evaluate how well a chatbot or AI assistant is performing. For example. My agent runs Gemini, and I am using OpenAI to assess it. This method takes a more advanced approach than a human one by asking an AI to “grade” another AI’s answers based on things like relevance, correctness, and clarity. It’s a nice way to test and improve generative AI systems in a more scalable way. Let’s quickly code an example. We will import the modules, create a prompt, a simple chat agent, and ask it about a description of the weather for the month of May in NYC. # Imports from agno.agent import Agent from agno.models.google import Gemini import os # Evaluation Modules from deepeval.test_case import LLMTestCase, LLMTestCaseParams from deepeval.metrics import GEval # Prompt prompt = "Describe the weather in NYC for May" # Create agent agent = Agent( model= Gemini(id="gemini-1.5-flash", api_key = os.environ.get("GEMINI_API_KEY")), description= "An assistant agent", instructions= ["Be sucint"], markdown= True, monitoring= True ) # Run agent response = agent.run(prompt) # Print response print(response.content) It responds: “Mild, with average highs in the 60s°F and lows in the 50s°F. Expect some rain“. Nice. Seems pretty good to me. But how can we put a number on it and show a potential manager or client how our agent is doing? Here is how: Create a test case passing the prompt and the response to the LLMTestCase class. Create a metric. We will use the method GEval and add a prompt for the model to test it for coherence, and then I give it the meaning of what coherence is to me. Give the output as evaluation_params. Run the measure method and get the score and reason from it. # Test Case test_case = LLMTestCase(input=prompt, actual_output=response) # Setup the Metric coherence_metric = GEval( name="Coherence", criteria="Coherence. The agent can answer the prompt and the response makes sense.", evaluation_params=[LLMTestCaseParams.ACTUAL_OUTPUT] ) # Run the metric coherence_metric.measure(test_case) print(coherence_metric.score) print(coherence_metric.reason) The output looks like this. 0.9 The response directly addresses the prompt about NYC weather in May, maintains logical consistency, flows naturally, and uses clear language. However, it could be slightly more detailed. 0.9 seems pretty good, given that the default threshold is 0.5. If you want to check the logs, use this next snippet. # Check the logs print(coherence_metric.verbose_logs) Here’s the response. Criteria: Coherence. The agent can answer the prompt and the response makes sense. Evaluation Steps: [ "Assess whether the response directly addresses the prompt; if it aligns, it scores higher on coherence.", "Evaluate the logical flow of the response; responses that present ideas in a clear, organized manner rank better in coherence.", "Consider the relevance of examples or evidence provided; responses that include pertinent information enhance their coherence.", "Check for clarity and consistency in terminology; responses that maintain clear language without contradictions achieve a higher coherence rating." ] Very nice. Now let us learn about another interesting use case, which is the evaluation of task completion for AI Agents. Elaborating a little more, how our agent is doing when it is requested to perform a task, and how much of it the agent can deliver. First, we are creating a simple agent that can access Wikipedia and summarize the topic of the query. # Imports from agno.agent import Agent from agno.models.google import Gemini from agno.tools.wikipedia import WikipediaTools import os from deepeval.test_case import LLMTestCase, ToolCall from deepeval.metrics import TaskCompletionMetric from deepeval import evaluate # Prompt prompt = "Search wikipedia for 'Time series analysis' and summarize the 3 main points" # Create agent agent = Agent( model= Gemini(id="gemini-2.0-flash", api_key = os.environ.get("GEMINI_API_KEY")), description= "You are a researcher specialized in searching the wikipedia.", tools= [WikipediaTools()], show_tool_calls= True, markdown= True, read_tool_call_history= True ) # Run agent response = agent.run(prompt) # Print response print(response.content) The result looks very good. Let’s evaluate it using the TaskCompletionMetric class. # Create a Metric metric = TaskCompletionMetric( threshold=0.7, model="gpt-4o-mini", include_reason=True ) # Test Case test_case = LLMTestCase( input=prompt, actual_output=response.content, tools_called=[ToolCall(name="wikipedia")] ) # Evaluate evaluate(test_cases=[test_case], metrics=[metric]) Output, including the agent’s response. ====================================================================== Metrics Summary - Task Completion (score: 1.0, threshold: 0.7, strict: False, evaluation model: gpt-4o-mini, reason: The system successfully searched for 'Time series analysis' on Wikipedia and provided a clear summary of the 3 main points, fully aligning with the user's goal., error: None) For test case: - input: Search wikipedia for 'Time series analysis' and summarize the 3 main points - actual output: Here are the 3 main points about Time series analysis based on the Wikipedia search: 1. **Definition:** A time series is a sequence of data points indexed in time order, often taken at successive, equally spaced points in time. 2. **Applications:** Time series analysis is used in various fields like statistics, signal processing, econometrics, weather forecasting, and more, wherever temporal measurements are involved. 3. **Purpose:** Time series analysis involves methods for extracting meaningful statistics and characteristics from time series data, and time series forecasting uses models to predict future values based on past observations. - expected output: None - context: None - retrieval context: None ====================================================================== Overall Metric Pass Rates Task Completion: 100.00% pass rate ====================================================================== ✓ Tests finished ! Run 'deepeval login' to save and analyze evaluation results on Confident AI. Our agent passed the test with honor: 100%! You can learn much more about the DeepEval library in this link [8]. Finally, in the next section, we will learn the capabilities of Agno’s library for monitoring agents. Agent Monitoring Like I told you in my previous post [9], I chose Agno to learn more about Agentic AI. Just to be clear, this is not a sponsored post. It is just that I think this is the best option for those starting their journey learning about this topic. So, one of the cool things we can take advantage of using Agno’s framework is the app they make available for model monitoring. Take this agent that can search the internet and write Instagram posts, for example. # Imports import os from agno.agent import Agent from agno.models.google import Gemini from agno.tools.file import FileTools from agno.tools.googlesearch import GoogleSearchTools # Topic topic = "Healthy Eating" # Create agent agent = Agent( model= Gemini(id="gemini-1.5-flash", api_key = os.environ.get("GEMINI_API_KEY")), description= f"""You are a social media marketer specialized in creating engaging content. Search the internet for 'trending topics about {topic}' and use them to create a post.""", tools=[FileTools(save_files=True), GoogleSearchTools()], expected_output="""A short post for instagram and a prompt for a picture related to the content of the post. Don't use emojis or special characters in the post. If you find an error in the character encoding, remove the character before saving the file. Use the template: - Post - Prompt for the picture Save the post to a file named 'post.txt'.""", show_tool_calls=True, monitoring=True) # Writing and saving a file agent.print_response("""Write a short post for instagram with tips and tricks that positions me as an authority in {topic}.""", markdown=True) To monitor its performance, follow these steps: Go to https://app.agno.com/settings and get an API Key. Open a terminal and type ag setup. If it is the first time, it might ask for the API Key. Copy and Paste it in the terminal prompt. You will see the Dashboard tab open in your browser. If you want to monitor your agent, add the argument monitoring=True. Run your agent. Go to the Dashboard on the web browser. Click on Sessions. As it is a single agent, you will see it under the tab Agents on the top portion of the page. Agno Dashboard after running the agent. Image by the author. The cools features we can see there are: Info about the model The response Tools used Tokens consumed This is the resulting token consumption while saving the file. Image by the author. Pretty neat, huh? This is useful for us to know where the agent is spending more or less tokens, and where it is taking more time to perform a task, for example. Well, let’s wrap up then. Before You Go We have learned a lot in this second round. In this post, we covered: Guardrails for AI are essential safety measures and ethical guidelines implemented to prevent unintended harmful outputs and ensure responsible AI behavior. Model evaluation, exemplified by GEval for broad assessment and TaskCompletion with DeepEval for agents output quality, is crucial for understanding AI capabilities and limitations. Model monitoring with Agno’s app, including tracking token usage and response time, which is vital for managing costs, ensuring performance, and identifying potential issues in deployed AI systems. Contact & Follow Me If you liked this content, find more of my work in my website. https://gustavorsantos.me GitHub Repository https://github.com/gurezende/agno-ai-labs References [1. Guardrails Ai] https://www.guardrailsai.com/docs/getting_started/guardrails_server [2. Guardrails AI Auth Key] https://hub.guardrailsai.com/keys [3. Guardrails AI Hub] https://hub.guardrailsai.com/ [4. Guardrails Restrict to Topic] https://hub.guardrailsai.com/validator/tryolabs/restricttotopic [5. DeepEval.] https://www.deepeval.com/docs/getting-started [6. DataCamp – DeepEval Tutorial] https://www.datacamp.com/tutorial/deepeval [7. DeepEval. TaskCompletion] https://www.deepeval.com/docs/metrics-task-completion [8. Llm Evaluation Metrics: The Ultimate LLM Evaluation Guide] https://www.confident-ai.com/blog/llm-evaluation-metrics-everything-you-need-for-llm-evaluation [9. Agentic AI 101: Starting Your Journey Building AI Agents] https://towardsdatascience.com/agentic-ai-101-starting-your-journey-building-ai-agents/ The post Agentic AI 102: Guardrails and Agent Evaluation appeared first on Towards Data Science.
    0 Comentários 0 Compartilhamentos 0 Anterior
  • Intel’s Spring Gaming Bundle On Amazon Slashes 33 Percent From The Incredibly Powerful Core Ultra 7 265K Desktop Processor, Now Available For A New Low $269.99

    Menu

    Home
    News

    Hardware

    Gaming

    Mobile

    Finance
    Deals
    Reviews
    How To

    Wccftech

    Deals
    Intel’s Spring Gaming Bundle On Amazon Slashes 33 Percent From The Incredibly Powerful Core Ultra 7 265K Desktop Processor, Now Available For A New Low Omar Sohail •
    May 17, 2025 at 07:17am EDT

    AMD might have surpassed Intel as the preferred brand for gaming desktop processors thanks to its ‘X3D’ range of CPUs, but with just enough of a discount, we believe that you can find a place in your heart for team blue once more. On Amazon, Intel’s Spring Gaming Bundle is live, and you can avail tons of offers on a variety of products, ranging from desktop CPUs, work laptops, gaming laptops, mini PCs, AIOs, and more. While doing our usual hunting around, our eyes immediately caught the Intel Core Ultra 7 265K.
    The high-end CPU is currently listed at 33 percent off, or a discount on the online retailer, bringing its price down to As numerous benchmarks would suggest, the Core Ultra 7 265K is not faster than AMD’s Ryzen 7 9800X3D, but at its updated price, it was never meant to compete with these chips, but provide consumers and creative professionals a more affordable alternative. You can simply save the money you would have otherwise invested in the Ryzen 7 9800X3D and repurpose that sum in a capable graphics card or any other component.

    The Intel Core Ultra 7 265K sports a 20-core hybrid architecture, with eight performance and 12 efficiency cores, and its clock speeds can reach up to 5.50GHz with the right amount of cooling. You get a decent 36MB of L3 cache, plus support for PCIe NVMe Gen 5 SSDs. Both gamers and creative professionals can leverage the extra core count for a number of applications, and only then will you start to notice the value of that 33 percent discount. We cannot confirm when Intel’s Spring Gaming Bundle will end, but you should be on top of this deal right away.
    Get the Intel Core Ultra 7 265K desktop CPU from Amazon -See more products from Intel’s Spring Gaming Bundle here
    See the latest technology deals that Amazon has discounted today
    See what else Amazon has discounted today

    Deal of the Day

    Subscribe to get an everyday digest of the latest technology news in your inbox

    Follow us on

    Topics

    Sections

    Company

    Some posts on wccftech.com may contain affiliate links. We are a participant in the Amazon Services LLC
    Associates Program, an affiliate advertising program designed to provide a means for sites to earn
    advertising fees by advertising and linking to amazon.com
    © 2025 WCCF TECH INC. 700 - 401 West Georgia Street, Vancouver, BC, Canada
    #intels #spring #gaming #bundle #amazon
    Intel’s Spring Gaming Bundle On Amazon Slashes 33 Percent From The Incredibly Powerful Core Ultra 7 265K Desktop Processor, Now Available For A New Low $269.99
    Menu Home News Hardware Gaming Mobile Finance Deals Reviews How To Wccftech Deals Intel’s Spring Gaming Bundle On Amazon Slashes 33 Percent From The Incredibly Powerful Core Ultra 7 265K Desktop Processor, Now Available For A New Low Omar Sohail • May 17, 2025 at 07:17am EDT AMD might have surpassed Intel as the preferred brand for gaming desktop processors thanks to its ‘X3D’ range of CPUs, but with just enough of a discount, we believe that you can find a place in your heart for team blue once more. On Amazon, Intel’s Spring Gaming Bundle is live, and you can avail tons of offers on a variety of products, ranging from desktop CPUs, work laptops, gaming laptops, mini PCs, AIOs, and more. While doing our usual hunting around, our eyes immediately caught the Intel Core Ultra 7 265K. The high-end CPU is currently listed at 33 percent off, or a discount on the online retailer, bringing its price down to As numerous benchmarks would suggest, the Core Ultra 7 265K is not faster than AMD’s Ryzen 7 9800X3D, but at its updated price, it was never meant to compete with these chips, but provide consumers and creative professionals a more affordable alternative. You can simply save the money you would have otherwise invested in the Ryzen 7 9800X3D and repurpose that sum in a capable graphics card or any other component. The Intel Core Ultra 7 265K sports a 20-core hybrid architecture, with eight performance and 12 efficiency cores, and its clock speeds can reach up to 5.50GHz with the right amount of cooling. You get a decent 36MB of L3 cache, plus support for PCIe NVMe Gen 5 SSDs. Both gamers and creative professionals can leverage the extra core count for a number of applications, and only then will you start to notice the value of that 33 percent discount. We cannot confirm when Intel’s Spring Gaming Bundle will end, but you should be on top of this deal right away. Get the Intel Core Ultra 7 265K desktop CPU from Amazon -See more products from Intel’s Spring Gaming Bundle here See the latest technology deals that Amazon has discounted today See what else Amazon has discounted today Deal of the Day Subscribe to get an everyday digest of the latest technology news in your inbox Follow us on Topics Sections Company Some posts on wccftech.com may contain affiliate links. We are a participant in the Amazon Services LLC Associates Program, an affiliate advertising program designed to provide a means for sites to earn advertising fees by advertising and linking to amazon.com © 2025 WCCF TECH INC. 700 - 401 West Georgia Street, Vancouver, BC, Canada #intels #spring #gaming #bundle #amazon
    WCCFTECH.COM
    Intel’s Spring Gaming Bundle On Amazon Slashes 33 Percent From The Incredibly Powerful Core Ultra 7 265K Desktop Processor, Now Available For A New Low $269.99
    Menu Home News Hardware Gaming Mobile Finance Deals Reviews How To Wccftech Deals Intel’s Spring Gaming Bundle On Amazon Slashes 33 Percent From The Incredibly Powerful Core Ultra 7 265K Desktop Processor, Now Available For A New Low $269.99 Omar Sohail • May 17, 2025 at 07:17am EDT AMD might have surpassed Intel as the preferred brand for gaming desktop processors thanks to its ‘X3D’ range of CPUs, but with just enough of a discount, we believe that you can find a place in your heart for team blue once more. On Amazon, Intel’s Spring Gaming Bundle is live, and you can avail tons of offers on a variety of products, ranging from desktop CPUs, work laptops, gaming laptops, mini PCs, AIOs, and more. While doing our usual hunting around, our eyes immediately caught the Intel Core Ultra 7 265K. The high-end CPU is currently listed at 33 percent off, or a $135 discount on the online retailer, bringing its price down to $269.99. As numerous benchmarks would suggest, the Core Ultra 7 265K is not faster than AMD’s Ryzen 7 9800X3D, but at its updated price, it was never meant to compete with these chips, but provide consumers and creative professionals a more affordable alternative. You can simply save the money you would have otherwise invested in the Ryzen 7 9800X3D and repurpose that sum in a capable graphics card or any other component. The Intel Core Ultra 7 265K sports a 20-core hybrid architecture, with eight performance and 12 efficiency cores, and its clock speeds can reach up to 5.50GHz with the right amount of cooling. You get a decent 36MB of L3 cache, plus support for PCIe NVMe Gen 5 SSDs. Both gamers and creative professionals can leverage the extra core count for a number of applications, and only then will you start to notice the value of that 33 percent discount. We cannot confirm when Intel’s Spring Gaming Bundle will end, but you should be on top of this deal right away. Get the Intel Core Ultra 7 265K desktop CPU from Amazon - $269.99 (33 percent off) See more products from Intel’s Spring Gaming Bundle here See the latest technology deals that Amazon has discounted today See what else Amazon has discounted today Deal of the Day Subscribe to get an everyday digest of the latest technology news in your inbox Follow us on Topics Sections Company Some posts on wccftech.com may contain affiliate links. We are a participant in the Amazon Services LLC Associates Program, an affiliate advertising program designed to provide a means for sites to earn advertising fees by advertising and linking to amazon.com © 2025 WCCF TECH INC. 700 - 401 West Georgia Street, Vancouver, BC, Canada
    0 Comentários 0 Compartilhamentos 0 Anterior
CGShares https://cgshares.com