https://
github.com/huggingface/trl + https://
huggingface.co/blog/async-rl-
training-landscape
… https://
huggingface.co/blog/unsloth-j
obs
… You'll have to create an account on HF to share your experiments btw!
OPEN SOURCE
-
Hugging Face TRL and Async RL Training Landscape Guide
By
–
-
Llama 31B Performance in llama-server: User Experience
By
–
Did you try the 31B in llama-server? Worked well for me
-
Google Gemma 4 31B achieves 2.7X faster performance on RTX
By
–
.@GoogleGemma 4 31B is up to 2.7X faster on RTX using llama.cpp. Thanks to @ggerganov for working with us to make this model fast.
→ View original post on X — @huggingface, 2026-04-02 19:30 UTC
-
Linux Kernel Quote Clarification OpenClaw Attribution
By
–
That quote is from the Linux Kernel, not OpenClaw.
-
Build Your Own AI to Protect Your Interests
By
–
Of course it will. But you will have your AI protect your interests. And bring you others. It really won't matter. Build your own AI.
-
Ollama Performance Comparable to Custom Compiled Alternatives
By
–
Actually I don't mind ollama (they'll probably have gemma 4 in a few days because the architecture is so similar to gemma 3 that it should be an easy port). I know I should consider other options, but somehow the performance is the same or better compared to custom compiled
-
Run Gemma 4 Locally in Browser Free Private
By
–
You can run Gemma 4 100% locally in your browser thanks to HF transformers.js. That means 100% private and 100% free! @xenovacom created a demo for it here: https://
huggingface.co/spaces/webml-c
ommunity/Gemma-4-WebGPU
… -
AI-Generated Security Reports Threaten Linux Kernel Projects
By
–
Prediction: This is gonna kill some oss projects. "On the kernel security list we've seen a huge bump of reports. We were between 2 and 3 per week maybe two years ago, then reached probably 10 a week over the last year with the only difference being only AI slop, and now since
-
DeepMind Launches Gemma 4: Open Source Models in Multiple Sizes
By
–
Really excited for this launch of Gemma 4 from @demishassabis and the DeepMind team. Open source models are a key front for the west to have a lead on and this is a very key addition to the effort. Excited to see what developers in SV and around the world can build using this. Demis Hassabis (@demishassabis) Excited to launch Gemma 4: the best open models in the world for their respective sizes. Available in 4 sizes that can be fine-tuned for your specific task: 31B dense for great raw performance, 26B MoE for low latency, and effective 2B & 4B for edge device use – happy building! — https://nitter.net/demishassabis/status/2039736628659269901#m
→ View original post on X — @demishassabis, 2026-04-02 19:07 UTC
-
Gemma 4 26B model architecture gallery entry and details
By
–
And the link to the gallery entry for more details, links, comparisons, etc: sebastianraschka.com/llm-arc…