Claude Opus 4.1 has about 2% better evals in the SWE benchmark. Moreover, it is already available. But pay even more attention to the comment afterwards: In the next few weeks (!) much bigger improvements will be released for their models. We are accelerating!
LLMS
-
Claude Opus 4.1 Release Announcement
By
–
Looks like Claude Opus 4.1 is already released!! https://
x.com/figuret20/stat
/figuret20/status/1952767582555238668
… -
Major Model Improvements Coming in Weeks Ahead
By
–
We plan to release substantially larger improvements to our models in the coming weeks.
-
Opus 4.1 Now Available to Claude Paid Users
By
–
Opus 4.1 is now available to paid Claude users and in Claude Code. It's also on our API, Amazon Bedrock, and Google Cloud's Vertex AI. Read more:
-
Claude Opus 4.1 Release: Agentic Tasks and Coding Improvements
By
–
Today we're releasing Claude Opus 4.1, an upgrade to Claude Opus 4 on agentic tasks, real-world coding, and reasoning.
-
Major AI Model Releases: Genie 3, GPT, and Claude 4.1
By
–
And another model: Claude opus 4.1: will Anthropic also drop a modal today?? What a day:
– Genie 3 from google deepmind
– OpenAI OSS GPT wie CUA
– Claude 4.1 -
OpenAI’s Open Source Model Now Features Computer Usage Capability
By
–
Holy moly, looks like OpenAI’s opensource model has computer usage https://t.co/tPUMb4JXZ3
— Chubby♨️ (@kimmonismus) 5 août 2025Holy moly, looks like OpenAI’s opensource model has computer usage
-
GPT-5 Coming This Week, Open Source Model Today
By
–
Big but small today – GPT-5 later this week – confirmed by Sams. So probably today the open source model!
-
How to Write Detailed AI Model Evaluation Articles
By
–
Is that how Microsoft Copilot does it? If you write that up in detail you will get SO much traffic to that article. Currently we are still living in a world where the only truly worthwhile articles on running evals come from @HamelHusain Someone should compete with him!
-
MacPaint Simulator Built with GLM-4.5 in 30 Seconds
By
–
vibe coding 𝗠𝗮𝗰𝗣𝗮𝗶𝗻𝘁 𝘄𝗶𝘁𝗵 𝗚𝗟𝗠-𝟰.𝟱 𝗶𝗻 𝗷𝘂𝘀𝘁 𝟯𝟬 𝘀𝗲𝗰𝗼𝗻𝗱𝘀 by @ting_
— AK (@_akhaliq) 5 août 2025
prompt: Make a Macintosh 128K simulator with MacPaint running in the virtual machine—it's usable, and users can draw on the canvas. pic.twitter.com/75oOymUFFUvibe coding 𝗠𝗮𝗰𝗣𝗮𝗶𝗻𝘁 𝘄𝗶𝘁𝗵 𝗚𝗟𝗠-𝟰.𝟱 𝗶𝗻 𝗷𝘂𝘀𝘁 𝟯𝟬 𝘀𝗲𝗰𝗼𝗻𝗱𝘀 by @ting_ prompt: Make a Macintosh 128K simulator with MacPaint running in the virtual machine—it's usable, and users can draw on the canvas.