what are some good evals for base models? will run! (FYI i expect it to not perform _that_ well since the model card emphasizes they really only pretrained gpt-oss to be good at math/code/reasoning. also my basemodelrecovery process is likely somewhat lossy.)
CODE
-
OpenAI Codex CLI Integration with Replicate MCP
By
–
OpenAI's Codex CLI works with Replicate MCP
— Replicate (@replicate) 13 août 2025
Use it to discover, compare, and run AI models from the comfort of your terminal.
Docs: https://t.co/j9D2zRO5YE pic.twitter.com/S1i4tvI6SYOpenAI's Codex CLI works with Replicate MCP Use it to discover, compare, and run AI models from the comfort of your terminal. Docs: https://
mcp.replicate.com/#codex -
Google Releases LangExtract Python Library for Data Extraction
By
–
Google just released a python library for data extraction! LangExtract is a python library that extracts structured information from unstructured text documents with precise source grounding and interactive visualization. 100% Open Source
-
Running AI Model Evaluations and Performance Benchmarks
By
–
can you give some examples? i bet it doesn't perform better– but i can run the evals!
-
Building AI Solutions Systematically: New Course Launch
By
–
We are focusing on building! Very little theory in this one. It's not about tricks and prompts; it's about building systematically and solving hard problems. This is useful for both beginners and experienced builders. The course launches tomorrow!
-
Mistral Medium 3.1 Now Available in Anycoder
By
–
Mistral Medium 3.1 is now available in anycoder as mistral-medium-2508
-
Mistral Medium 3.1 Available in Anycoder Platform
By
–
Mistral Medium 3.1 is now available in anycoder as mistral-medium-2508
-
GPT-5-Pro Download Link Generation Issues and Code Display
By
–
I had several issues with GPT-5-Pro generating a downloadable link (e.g. HTML file) and that link wasn't working – just showing a download error (not the 'code interpreter sessions expired' one – this is separate). So I was asking it to show generated code as well, that worked,
-
GPT-5: AI Automation for Code, Scheduling, and Emails
By
–
Tight deadlines. Endless meetings. Constant context-switching. Now imagine an AI that:
Analyzes your bottlenecks Generates & debugs code Automates scheduling & emails
That’s GPT-5. Here’s how it works https://
tinyurl.com/42r7auwx courtesy of http://
Futurepedia.io – your -
Vercel launches v0.app agentic AI web developer for everyone
By
–
Vercel updated https://t.co/qoXY36qeXl to https://t.co/4H7pxmxpjQ, an agentic web app developer for everyone
— The Rundown AI (@TheRundownAI) 13 août 2025
The revamped offering uses agentic AI to plan, research, build, and debug, while working in context, across multiple steps, and adapting on the gopic.twitter.com/pn22AakMCRVercel updated http://
v0.dev to http://
v0.app, an agentic web app developer for everyone The revamped offering uses agentic AI to plan, research, build, and debug, while working in context, across multiple steps, and adapting on the go