正在加载视频...

视频加载失败

加载此视频时出现问题。这可能是由于临时网络问题，或视频可能不可用。

Orchard is an open-source framework for the research community to train and evaluate AI agents across task types. It reduces complexity while supporting strong performance from smaller models by enabling researchers to reuse the same infrastructure.

Microsoft Research

554,722 subscribers

23,308 次观看 • 7 天前 •via X (Twitter)

Anya Rossi• Live Now

Private livecam show

0 条评论

暂无评论

原始帖子的评论将显示在这里

相关视频

Super excited to share 🧠MLGym 🦾 – the first Gym environment for AI Research Agents 🤖🔬 We introduce MLGym and MLGym-Bench, a new framework and benchmark for evaluating and developing LLM agents on AI research tasks. The key contributions of our work are: 🕹️ Enables the exploration of different training algorithms for AI Research Agents such as RL 🛠️ Provides a flexible evaluation framework that can accommodate different artifacts such as models, algorithms, or predictions 🤖 Allows researchers to evaluate any model without the need to develop a custom agentic harness 🎯 Introduces 13 diverse open-ended AI Research tasks for evaluating AI Research Agents on a wide range of domains such as computer vision, natural language processing, reinforcement learning, game theory, and logical reasoning. 📈 Proposes a new evaluation metric for AI Research Agents MLGym makes it easy to: 1) Add new tasks 2) Evaluate new models 3) Integrate new agents Check out a video of the MLGym Agent to see how it performs the full pipeline of idea generation💡, implementation 👩‍💻, experimentation 👩‍🔬, and iteration 🔄 to improve on ML tasks. Huge thanks to the exceptionally talented Deepak Nathani who led this work and to all the other amazing collaborators who made this possible 🙏🫶🚀

Super excited to share 🧠MLGym 🦾 – the first Gym environment for AI Research Agents 🤖🔬 We introduce MLGym and MLGym-Bench, a new framework and benchmark for evaluating and developing LLM agents on AI research tasks. The key contributions of our work are: 🕹️ Enables the exploration of different training algorithms for AI Research Agents such as RL 🛠️ Provides a flexible evaluation framework that can accommodate different artifacts such as models, algorithms, or predictions 🤖 Allows researchers to evaluate any model without the need to develop a custom agentic harness 🎯 Introduces 13 diverse open-ended AI Research tasks for evaluating AI Research Agents on a wide range of domains such as computer vision, natural language processing, reinforcement learning, game theory, and logical reasoning. 📈 Proposes a new evaluation metric for AI Research Agents MLGym makes it easy to: 1) Add new tasks 2) Evaluate new models 3) Integrate new agents Check out a video of the MLGym Agent to see how it performs the full pipeline of idea generation💡, implementation 👩‍💻, experimentation 👩‍🔬, and iteration 🔄 to improve on ML tasks. Huge thanks to the exceptionally talented Deepak Nathani who led this work and to all the other amazing collaborators who made this possible 🙏🫶🚀

Roberta Raileanu

105,052 次观看 • 1 年前

We're excited to announce that Coinbase Developer Platform🛡️ AgentKit supports the new OpenAI Agents SDK. OpenAI’s new Agents SDK is an open source framework for building and scaling agents that includes built-in tools for web search, file search, and computer use, and tools to track and optimize agent performance—making it easier than ever to build production-ready AI agents. Coinbase AgentKit complements this by adding secure crypto wallets directly to your agents, enabling them to transact globally, instantly, and with near-zero fees—unlocking true financial autonomy.

We're excited to announce that Coinbase Developer Platform🛡️ AgentKit supports the new OpenAI Agents SDK. OpenAI’s new Agents SDK is an open source framework for building and scaling agents that includes built-in tools for web search, file search, and computer use, and tools to track and optimize agent performance—making it easier than ever to build production-ready AI agents. Coinbase AgentKit complements this by adding secure crypto wallets directly to your agents, enabling them to transact globally, instantly, and with near-zero fees—unlocking true financial autonomy.

Coinbase Developer Platform🛡️

20,287 次观看 • 1 年前

Introducing web-agent, an open framework for building web agents 🔥 Build AI agents that search, scrape, and interact with the web - powered by the same architecture behind our /agent endpoint. 100% open source. Bring any model. Anthropic, OpenAI, or your own.

Introducing web-agent, an open framework for building web agents 🔥 Build AI agents that search, scrape, and interact with the web - powered by the same architecture behind our /agent endpoint. 100% open source. Bring any model. Anthropic, OpenAI, or your own.

Firecrawl

123,675 次观看 • 3 个月前

Top announcement from the Developer Keynote: Gemini Enterprise Agent Platform—a unified dev ecosystem for building production-ready AI agents and infrastructure. What's that mean? Access to 200+ models, including Gemini and Gemma 4 + open source models →

Top announcement from the Developer Keynote: Gemini Enterprise Agent Platform—a unified dev ecosystem for building production-ready AI agents and infrastructure. What's that mean? Access to 200+ models, including Gemini and Gemma 4 + open source models →

Google Cloud Tech

16,489 次观看 • 3 个月前

To scale Perplexity, CEO Aravind Srinivas says large models now label, evaluate, and train smaller ones. The system handles millions of queries daily. Human evaluation is no longer the bottleneck. “An AI has to do this job.”

To scale Perplexity, CEO Aravind Srinivas says large models now label, evaluate, and train smaller ones. The system handles millions of queries daily. Human evaluation is no longer the bottleneck. “An AI has to do this job.”

vitrupo

61,760 次观看 • 1 年前

Thrilled to see Amazon Web Services making a major contribution to the open source AI community with the launch of the Strands Agents, an open source AI agents SDK! The core of Strands is the simple agentic loop that connects the model and tools together, like the two strands of DNA. This model-driven approach to agent building eliminates the need for complex agent orchestration by embracing the capabilities of state-of-the-art models to plan, chain thoughts, call tools, and reflect. Providing open source tools and interoperability with open source protocols is an important part of our strategy to enable an agentic future. Can't wait to see what you build with Strands!

Thrilled to see Amazon Web Services making a major contribution to the open source AI community with the launch of the Strands Agents, an open source AI agents SDK! The core of Strands is the simple agentic loop that connects the model and tools together, like the two strands of DNA. This model-driven approach to agent building eliminates the need for complex agent orchestration by embracing the capabilities of state-of-the-art models to plan, chain thoughts, call tools, and reflect. Providing open source tools and interoperability with open source protocols is an important part of our strategy to enable an agentic future. Can't wait to see what you build with Strands!

Swami Sivasubramanian

32,185 次观看 • 1 年前

The AI fun continues ✨ You can use Unity Sentis in open beta to build AI-powered experiences in the Unity runtime. Sentis reduces the complexity of importing and running AI models and enables creating lifelike NPCs and new interactions for richer gameplay. See it in the demo below that uses four chained AI models to create an open-ended dialogue and quest experience. Learn more about Sentis: #Unite2023 #UnitySentis

The AI fun continues ✨ You can use Unity Sentis in open beta to build AI-powered experiences in the Unity runtime. Sentis reduces the complexity of importing and running AI models and enables creating lifelike NPCs and new interactions for richer gameplay. See it in the demo below that uses four chained AI models to create an open-ended dialogue and quest experience. Learn more about Sentis: #Unite2023 #UnitySentis

Unity

62,681 次观看 • 2 年前

💡 Acronym of the day: EGI [Enterprise General Intelligence] Salesforce AI Research coined it to define how well AI agents deliver consistent, capable performance across business tasks. 🔎 Dig deeper into our new framework for excellence in AI systems for business:

💡 Acronym of the day: EGI [Enterprise General Intelligence] Salesforce AI Research coined it to define how well AI agents deliver consistent, capable performance across business tasks. 🔎 Dig deeper into our new framework for excellence in AI systems for business:

Salesforce

30,804 次观看 • 1 年前

Cua (Cua) is the Docker for computer-use agents, an open-source framework that enables AI agents to control full operating systems within lightweight virtual containers, and works with any language model. Congrats on the launch, Francesco + Sandro!

Cua (Cua) is the Docker for computer-use agents, an open-source framework that enables AI agents to control full operating systems within lightweight virtual containers, and works with any language model. Congrats on the launch, Francesco + Sandro!

Y Combinator

105,618 次观看 • 1 年前

Meta PARTNR is a research framework supporting seamless human-robot collaboration. Building on our research with Habitat, we’re open sourcing a large-scale benchmark, dataset and large planning model that we hope will enable the community to effectively train social robots.

Meta PARTNR is a research framework supporting seamless human-robot collaboration. Building on our research with Habitat, we’re open sourcing a large-scale benchmark, dataset and large planning model that we hope will enable the community to effectively train social robots.

AI at Meta

106,951 次观看 • 1 年前

AI creators now have the economic infrastructure to build, deploy, and earn; it's 100% open-source and community-driven. OpenLedger offers unmatched transparency, flexibility, and the tools you need to build the future of AI on your terms. OpenLedger is a decentralized, open-source blockchain designed for sourcing, training, and monetizing AI models. Learn more about in this video 👇

AI creators now have the economic infrastructure to build, deploy, and earn; it's 100% open-source and community-driven. OpenLedger offers unmatched transparency, flexibility, and the tools you need to build the future of AI on your terms. OpenLedger is a decentralized, open-source blockchain designed for sourcing, training, and monetizing AI models. Learn more about in this video 👇

MosesMee

14,108 次观看 • 1 年前

"Security by obscurity is going to zero rapidly." James O'Beirne and other researchers independently reproduced the COLDCARD vulnerability by pointing an AI model at the firmware history. This is the new reality for open-source security.

"Security by obscurity is going to zero rapidly." James O'Beirne and other researchers independently reproduced the COLDCARD vulnerability by pointing an AI model at the firmware history. This is the new reality for open-source security.

TFTC

13,879 次观看 • 10 天前

Z.ai GLM 5.2 is now live on Eigent from day 0! threw a real long-horizon task: research 30 companies across 6 sectors of the AI infrastructure stack, structure it into JSON, then build an interactive HTML report. same prompt, 5.1 vs 5.2. where 5.2 pulls ahead: → plans deeper and verifies harder. → researches wider. Covers 2x the ground. → ships a more complete, interactive report. with a 1M context window, it holds the whole task from research to final deliverable. no dropped threads, no half-finished output. long tasks are where the difference shows. try it on open source coworker Eigent. 🦾

Z.ai GLM 5.2 is now live on Eigent from day 0! threw a real long-horizon task: research 30 companies across 6 sectors of the AI infrastructure stack, structure it into JSON, then build an interactive HTML report. same prompt, 5.1 vs 5.2. where 5.2 pulls ahead: → plans deeper and verifies harder. → researches wider. Covers 2x the ground. → ships a more complete, interactive report. with a 1M context window, it holds the whole task from research to final deliverable. no dropped threads, no half-finished output. long tasks are where the difference shows. try it on open source coworker Eigent. 🦾

Eigent AI

22,881 次观看 • 1 个月前

We're excited to announce Hyper-Research. Open-source tool for comparing any AI models, products, or research, the same engine we used to stack Cursor, Antigravity, and Windsurf side-by-side. • Drop URLs • Get instant comparisons • Scoring dashboards + visual diffs • Powered by Hyperbrowser + Claude Opus 4.5

We're excited to announce Hyper-Research. Open-source tool for comparing any AI models, products, or research, the same engine we used to stack Cursor, Antigravity, and Windsurf side-by-side. • Drop URLs • Get instant comparisons • Scoring dashboards + visual diffs • Powered by Hyperbrowser + Claude Opus 4.5

Hyperbrowser

11,338 次观看 • 8 个月前

Robotics: coding agents’ next frontier. So how good are they? We introduce CaP-X: an open-source framework and benchmark for coding agents, where they write code for robot perception and control, execute it on sim and real robots, observe the outcomes, and iteratively improve code reliability. From NVIDIA Berkeley AI Research CMU Robotics Institute Stanford AI Lab 🧵

Robotics: coding agents’ next frontier. So how good are they? We introduce CaP-X: an open-source framework and benchmark for coding agents, where they write code for robot perception and control, execute it on sim and real robots, observe the outcomes, and iteratively improve code reliability. From NVIDIA Berkeley AI Research CMU Robotics Institute Stanford AI Lab 🧵

Max Fu

175,629 次观看 • 4 个月前

Open science is how we continue to push technology forward and today at Meta FAIR we’re sharing eight new AI research artifacts including new models, datasets and code to inspire innovation in the community. More in the video from Joelle Pineau. This work is another important step towards our goal of achieving Advanced Machine Intelligence (AMI). What we’re releasing: • Meta Spirit LM: An open source language model for seamless speech and text integration. • Meta Segment Anything Model 2.1: An updated checkpoint with improved results on visually similar objects, small objects and occlusion handling. Plus a new developer suite to make it easier for developers to build with SAM 2. • Layer Skip: Inference code and fine-tuned checkpoints demonstrating a new method for enhancing LLM performance. • SALSA: New code to enable researchers to benchmark AI-based attacks in support of validating security for post-quantum cryptography. • Meta Lingua: A lightweight and self-contained codebase designed to train language models at scale. • Meta Open Materials: New open source models and the largest dataset of its kind to accelerate AI-driven discovery of new inorganic materials. • MEXMA: A new research paper and code for our novel pre-trained cross-lingual sentence encoder with coverage across 80 languages. • Self-Taught Evaluator: a new method for generating synthetic preference data to train reward models without relying on human annotations. Access to state-of-the-art AI creates opportunities for everyone. We’re excited to share this work and look forward to seeing the community innovation that results from it. Details and access to everything released by FAIR today ➡️

Open science is how we continue to push technology forward and today at Meta FAIR we’re sharing eight new AI research artifacts including new models, datasets and code to inspire innovation in the community. More in the video from Joelle Pineau. This work is another important step towards our goal of achieving Advanced Machine Intelligence (AMI). What we’re releasing: • Meta Spirit LM: An open source language model for seamless speech and text integration. • Meta Segment Anything Model 2.1: An updated checkpoint with improved results on visually similar objects, small objects and occlusion handling. Plus a new developer suite to make it easier for developers to build with SAM 2. • Layer Skip: Inference code and fine-tuned checkpoints demonstrating a new method for enhancing LLM performance. • SALSA: New code to enable researchers to benchmark AI-based attacks in support of validating security for post-quantum cryptography. • Meta Lingua: A lightweight and self-contained codebase designed to train language models at scale. • Meta Open Materials: New open source models and the largest dataset of its kind to accelerate AI-driven discovery of new inorganic materials. • MEXMA: A new research paper and code for our novel pre-trained cross-lingual sentence encoder with coverage across 80 languages. • Self-Taught Evaluator: a new method for generating synthetic preference data to train reward models without relying on human annotations. Access to state-of-the-art AI creates opportunities for everyone. We’re excited to share this work and look forward to seeing the community innovation that results from it. Details and access to everything released by FAIR today ➡️

AI at Meta

150,222 次观看 • 1 年前

AI is undeniably a positive force, but it's up to us to push research forward ethically and responsibly. For this reason, we've partnered with an incredible group of researchers to develop CodeCarbon, an open source library to track and reduce CO2 emissions from your computing

AI is undeniably a positive force, but it's up to us to push research forward ethically and responsibly. For this reason, we've partnered with an incredible group of researchers to develop CodeCarbon, an open source library to track and reduce CO2 emissions from your computing

Comet

7,781,681 次观看 • 2 年前

Tuning AI models is an increasingly important task — and where DePIN thrives. As Emad discussed at #RenderCon, the future of AI relies on distributed compute platforms like The Render Network to tune, scale, and personalize agents across industries.

Tuning AI models is an increasingly important task — and where DePIN thrives. As Emad discussed at #RenderCon, the future of AI relies on distributed compute platforms like The Render Network to tune, scale, and personalize agents across industries.

The Render Network

16,764 次观看 • 1 年前