Move to Linux and try LMStudio and then Exl2/Exl3
OPEN SOURCE
-
24B Model Running in Browser with WebGPU Technology
By
–
24B model running in your browser! https://t.co/dorx4Ux2x7
— Maxime Labonne (@maximelabonne) 25 mars 202624B model running in your browser! Xenova (@xenovacom) WebGPU is INSANE! π€― Here's a 24B parameter model running locally in a web browser, at a blazing ~50 tokens/second on my M4 Max. β‘οΈ It's the largest model we've ever run with Transformers.js… and we're not stopping here. Big announcement soon. β https://nitter.net/xenovacom/status/2036908326462665211#m
β View original post on X β @maximelabonne, 2026-03-25 21:10 UTC
-
Cohere Defines the Future of Sovereign AI with Full Stack
By
–
.@Cohere is defining the future of sovereign AI.
— NVIDIA AI (@NVIDIAAI) 25 mars 2026
Autumn Moulder, VP of Engineering at Cohere, breaks down the requirements for a full AI sovereign stack at #NVIDIAGTC:
β Full-Stack Sovereignty: Deploying everything from the model to applications and reasoning traces within aβ¦ pic.twitter.com/i5vb5NSIqZ@Cohere is defining the future of sovereign AI. Autumn Moulder, VP of Engineering at Cohere, breaks down the requirements for a full AI sovereign stack at #NVIDIAGTC: β Full-Stack Sovereignty: Deploying everything from the model to applications and reasoning traces within a single data center. β The Power of Openness: Why open models like NVIDIA Nemotron are critical for data lineage and navigating evolving regulations. [Translated from EN to English]
β View original post on X β @nvidiaai, 2026-03-25 21:00 UTC
-
Codex App Server Open Source Platform for ChatGPT Integration
By
–
Your periodic reminder: Codex App Server is 100% Open Source!! It allows you to build on codex and build richer experiences all with ChatGPT Auth Thousands oof projects are built on top of it including our own VS Code Extension – check it out π
-
ARC-AGI-3 Benchmark Results and Saturation Timeline
By
–
ARC-AGI-3 benchmark is here It took from $2 to $9k for frontier models to complete the task at 0.2-0.3% acheivemnt. How soon would you expect it to get saturated?
-
Jensen Huang Discusses Open Frontier Models with AI Leaders
By
–
Different voices. Same answer: open models.
— NVIDIA (@nvidia) 25 mars 2026
NVIDIA Founder and CEO Jensen Huang sat down with the leaders from @mistralai, @bfl_ml, @cursor_ai, @LangChain, @perplexity_ai, @reflection_ai, @thinkymachines, @allen_ai, @evidenceopen, and AMP PBC to discuss the rapid rise of open⦠pic.twitter.com/wha1gE3W5oDifferent voices. Same answer: open models. NVIDIA Founder and CEO Jensen Huang sat down with the leaders from @mistralai, @bfl_ml, @cursor_ai, @LangChain, @perplexity_ai, @reflection_ai, @thinkymachines, @allen_ai, @evidenceopen, and AMP PBC to discuss the rapid rise of open frontier models. Get the top takeaways from the frontier of AI: nvda.ws/4uR6eEV
β View original post on X β @nvidia, 2026-03-25 19:03 UTC
-
HF Papers CLI: Semantic Search Tool for AI Agents on arXiv
By
–
HF Papers is the biggest infra for AI agents to do retrieval over arxiv
— AK (@_akhaliq) 25 mars 2026
introducing ππ ππππππ cli so that autoresearch can do semantic search & markdown retrieval of papers
ππ ππππππ [ππππππ, ππππ] pic.twitter.com/cTWM4GO9E6HF Papers is the biggest infra for AI agents to do retrieval over arxiv introducing ππ ππππππ cli so that autoresearch can do semantic search & markdown retrieval of papers ππ ππππππ [ππππππ, ππππ]
β View original post on X β @julien_c, 2026-03-25 17:07 UTC
-
P-video Speed Boost: 720p Videos in 5 Seconds for $0.02
By
–
β‘οΈP-video is faster now. A 720p 5-second video takes only 5 seconds at 0.02$ per second!
— Pruna AI (@PrunaAI) 25 mars 2026
Sign up for our API now: https://t.co/eZG4TRw9Ge
Try our Endpoint Guides: https://t.co/15z77wdOEU
Star our OSS: https://t.co/nLypK0Shkr pic.twitter.com/WEGQ7t87KHβ‘οΈP-video is faster now. A 720p 5-second video takes only 5 seconds at 0.02$ per second! Sign up for our API now: dashboard.pruna.ai/login Try our Endpoint Guides: docs.pruna.ai/en/stable/docsβ¦ Star our OSS: github.com/PrunaAI/pruna
β View original post on X β @replicate, 2026-03-25 16:01 UTC
-
Open Source and Science Research Experiments in AI
By
–
Let's gooo! excited to see more open source/ science experiments and research from you!
-
Broken retrieval systems can’t be fixed with prompts
By
–
Hard truth for every AI builder: you cannot prompt your way out of a broken retrieval system.
— God of Prompt (@godofprompt) 25 mars 2026
You've probably experienced this. You write the perfect system prompt. You fine-tune the agent instructions. But the agent still fires on the wrong context because the context it was⦠https://t.co/nlxImXOwqZ pic.twitter.com/tTh1nHgAETHard truth for every AI builder: you cannot prompt your way out of a broken retrieval system. You've probably experienced this. You write the perfect system prompt. You fine-tune the agent instructions. But the agent still fires on the wrong context because the context it was