Metis PCIe driver source is now on GitHub The metis.ko kernel module, GPL-2.0 licensed. Useful if you're doing custom kernel builds, Yocto, or non-standard Linux setups. Most users won't need it (Voyager SDK handles installation), but if you're experimenting, it's all there.
SOFTWARE
-
ContextualAI’s Agent Composer kills n8n, awesome
By
–
Reading about what @ContextualAI built with Agent Composer This completely kills n8n – awesome stuff.
-
Scobleizer: Typeless and Whisper Are Transformational Technologies for iPhone
By
–
I'm using Typeless, a magical tool that's the same kind of thing as Whisper. I'm also trying Whisper, and it's also good. These are transformational technologies for your iPhone if you start using them!
-
AI video editing cuts hours to minutes with equal quality
By
–
2 hours editing one video vs 12 minutes.@Roboneo_ai just made video editing look stupid.
— God of Prompt (@godofprompt) 29 janvier 2026
Tested it on a product launch. Same quality. Didn't sacrifice shit.
This is genuinely insane. pic.twitter.com/c9TT5np3dW2 hours editing one video vs 12 minutes. @Roboneo_ai just made video editing look stupid. Tested it on a product launch. Same quality. Didn't sacrifice shit. This is genuinely insane.
-
vLLM v0.14.0: Scheduler bug fix for Mamba token allocation
By
–
4/5 The confusion: Scheduler saw "1 token allocated" and assumed decode. But 1 token can also mean "new request that hit token budget limits." Now fixed: check whether it's the request's first-ever token, not the allocation size. In vLLM v0.14.0 – update if running Mamba.
-
Investigation: From Confident Gibberish to Scheduler Fix
By
–
5/5 Wrote up the complete investigation-from confident gibberish to the scheduler fix. Includes the false starts (suspected CUDA bugs), the breakthrough (request tracking), and lessons for anyone running stateful models at scale. Read the blog:
-
GPU Memory Utilization Bug: Token Classification Mismatch
By
–
2/5 Used gpu_memory_utilization=0.2 to reproduce quickly, but this happens naturally when the scheduler runs out of token budget and GPU blocks get recycled. New request gets 1 token → misclassified as "decode" But num_computed_tokens=0 → should be "prefill".
-
Custom Software Development at OpenAI with AI Tools
By
–
"Why do I buy planning software off the shelf?
Why wouldn't I code the kind of software that is exactly what OpenAl needs?" "OpenAl's finance team is just 18% the size of comparable companies. Why? Because their employees are building their own too with AI." "Building bespoke -
Google Adds CI Fixer and New MCP Integrations
By
–
Google released MCP integrations and a new "CI Fixer" feature in Beta for Jules SWE Agent.
— 🚨 AI News | TestingCatalog (@testingcatalog) 29 janvier 2026
CI Fixer "When enabled, Jules will automatically attempt to fix failing CI checks on pull requests it creates and push updates."
New MCPs include Linear, New Relic, Supabase, Neon,… pic.twitter.com/kKo8H9aGmCGoogle released MCP integrations and a new "CI Fixer" feature in Beta for Jules SWE Agent. CI Fixer "When enabled, Jules will automatically attempt to fix failing CI checks on pull requests it creates and push updates." New MCPs include Linear, New Relic, Supabase, Neon,
-
Gemini Chrome Side Panel Update Enhances Multitasking Efficiency
By
–
We’re introducing a series of updates that make Gemini in @googlechrome more helpful, efficient, and personalized for you:
— Google AI (@GoogleAI) 28 janvier 2026
— Side panel for multitasking: The new side panel experience allows you to do things like send emails or compare information without leaving your open tab… pic.twitter.com/5YY9yrhmb0We’re introducing a series of updates that make Gemini in @googlechrome more helpful, efficient, and personalized for you: — Side panel for multitasking: The new side panel experience allows you to do things like send emails or compare information without leaving your open tab