Results for “agents”

How Computer-Use Agents See, Act and Recover Ari Weinstein and Nikunj Handa explain the interface and API choices behind faster agents, and where their limits remain.

[AINews] Cursor 2.0 & Composer-1: Fast Models and New Agents UI Cursor 2.0 launched with Composer-1, an agentic coding model optimized for speed and precision, featuring multi-agent orchestration, built-in browser for testing, and…

ICLR 2024, Part 1 of 5: Finding the Tasks Agents Still Cannot Do Part 1 follows the path from rebuilding an NLP course after ChatGPT to web agents, social evaluation, code optimization and the compute constraints shaping academic research.

How Anthropic’s Dynamic Workflows Coordinate Claude Agents (Updated with video!) At AI Engineer New York, Anthropic drew on lessons from Unix and Linux to explain how agents can tackle bigger projects without overwhelming developers.

Mistral’s path from streaming speech to verifiable agents Voxtral TTS generates speech one frame at a time. The same conversation connects specialized models, enterprise training, formal proofs, and learning from agents that work for hours.

[AINews] Adept Fuyu-Heavy: Multimodal model for Agents Adept launched Fuyu-Heavy, a multimodal model focused on UI understanding and visual QA, outperforming Gemini Pro on the MMMU benchmark. The model size is estimated between 20B-170B…

Multi-Turn RL: Rewarding Agents for Useful Work From coding agents that do too much to tools that get called but ignored, the training problem is deciding which actions deserve credit.

[AINews] Claude Agent Skills - glorified AGENTS.md? or MCP killer? Anthropic achieves a rare feat with back-to-back AI news headlines featuring Claude's new Skills—a novel way to build specialized agents using Markdown files, scripts, and…

OpenAI’s Agents Platform: From Tool Calls to Observable Workflows The March 2025 launch conversation explains why Responses became a new API, how its tools compose, and how handoffs and traces could support better agents.

[AINews] Andrew likes Agents Andrew Ng's The Batch highlighted the significant improvement in code generation performance when using an iterative agent workflow, with GPT-3.5 wrapped in an agent loop achieving up to 95.1% accuracy on the…

Red teaming agents: from malicious inputs to consequential actions Gray Swan connects adversarial testing, specialized guard models and system permissions to make agents safer without stripping away their usefulness.

…Vibe Coding Manifesto: From Writing Code to Coordinating Agents Coding agents move the work toward asking questions, supervising execution, and integrating changes. The promise comes with a learning curve—and a bottleneck that more code…