How we built Claude Code auto mode: a safer way to skip permissions
Claude Code users approve 93% of permission prompts. We built classifiers to automate some decisions, increasing safety while reducing approval fatigue. Here's what it catches, and…
Frontier Labs 636 stories
Harness design for long-running application development
Harness design is key to performance at the frontier of agentic coding. Here's how we pushed Claude further in frontend design and long-running autonomous software engineering.
Speaking of Voxtral
Voxtral TTS: A frontier, open-weights text-to-speech model that’s fast, instantly adaptable, and produces lifelike speech for voice agents.
How we monitor internal coding agents for misalignment
How we monitor internal coding agents for misalignment
Introducing Imagine Gallery
Character.ai’s Imagine feature turns conversations into visuals - dramatic scenes, surreal worlds, unexpected moments. It’s one of the most creative ways to build upon your chat…
Introducing Forge
Today, we’re introducing Forge, a system that allows enterprises to build frontier-grade AI models grounded in their proprietary knowledge.
Cohere advances sovereign AI capabilities with NVIDIA
NVIDIA ecosystem-native model and local North instance will meet growing market demand for secure, privately run AI systems
Mistral AI partners with NVIDIA to accelerate open frontier models
As a founding member of the NVIDIA Nemotron Coalition, Mistral AI is contributing large-scale model development and multimodal capabilities.
Leanstral: Open-Source foundation for trustworthy vibe-coding
First open-source code agent for Lean 4.
Rails testing on autopilot: Building an agent that writes what developers won't
Rails testing on autopilot: Building an agent that writes what developers won't
Sandbox API: Isolated Code Execution for AI Agents
Sandbox API: Isolated Code Execution for AI Agents
Four MTIA Chips in Two Years: Scaling AI Experiences for Billions
Four MTIA Chips in Two Years: Scaling AI Experiences for Billions
Mapping the World's Forests with Greater Precision: Introducing Canopy Height Maps v2
Today, in partnership with the World Resources Institute, we’re announcing Canopy Height Maps v2 (CHMv2): an open source model and world-scale maps generated with it.
Eval awareness in Claude Opus 4.6’s BrowseComp performance
Evaluating Opus 4.6 on BrowseComp, we found cases where the model recognized the test, then found and decrypted answers to it—raising questions about eval integrity in…
Reverse engineering Claude's CVE-2026-2796 exploit
This post dives deep into how Claude wrote an exploit for one of the vulnerabilities it found in Firefox.
Partnering with Mozilla to improve Firefox’s security
In a collaboration with researchers at Mozilla, Claude Opus 4.6 discovered 22 Firefox vulnerabilities over the course of two weeks.
Where things stand with the Department of War
AnnouncementsPolicyWhere things stand with the Department of WarMar 5, 2026A statement from Dario AmodeiYesterday (March 4) Anthropic received a letter from the Department of War confirming…
Reasoning models struggle to control their chains of thought, and that’s good
Reasoning models struggle to control their chains of thought, and that’s good
No Cloud, No Waiting: Tool-Calling Agents on Consumer Hardware with LFM2-24B-A2B | Liquid AI
Building a local AI agent sounds great until you try to use one all day. The hard part isn’t getting a model to understand you, it’s…
Statement on the comments from Secretary of War Pete Hegseth
AnnouncementsPolicyStatement on the comments from Secretary of War Pete HegsethFeb 27, 2026Earlier today, Secretary of War Pete Hegseth shared on X that he is directing the…
Statement from Dario Amodei on our discussions with the Department of War
AnnouncementsPolicyStatement from Dario Amodei on our discussions with the Department of War Feb 26, 2026I believe deeply in the existential importance of using AI to defend…
From Cloud to AI PC: Launching Liquid’s Largest LFM2 Model Alongside Our Growing Partner Ecosystem | Liquid AI
Today, we are completing the LFM2 family with the launch of our most capable model yet: LFM2-24B-A2B. While our Technical Blog dives into the architectural specs,…
LFM2-24B-A2B: Scaling Up the LFM2 Architecture | Liquid AI
Today, we release an early checkpoint of LFM2-24B-A2B, our largest LFM2 model. This sparse Mixture of Experts (MoE) model has 24 billion total parameters with 2…
Cohere Labs Launches Tiny Aya, Making Multilingual AI Accessible
Cohere Labs Launches Tiny Aya, Making Multilingual AI Accessible