view article Article GGML and llama.cpp join HF to ensure the long-term progress of Local AI +4 ggerganov, ngxson, allozaur, lysandre, victor, julien-c • Feb 20 • 505
view article Article Welcome Gemma 4: Frontier multimodal intelligence on device +5 merve, pcuenq, sergiopaniego, burtenshaw, Steveeeeeeen, alvarobartt, SaylorTwift • Apr 2 • 890
view article Article Liberate your OpenClaw +6 clem, burtenshaw, pcuenq, jeffboudier, merve, nielsr, victor, mishig • Mar 27 • 45
view article Article Introducing Modular Diffusers - Composable Building Blocks for Diffusion Pipelines +2 YiYiXu, OzzyGT, dn6, sayakpaul • Mar 5 • 51
view article Article Introducing Storage Buckets on the Hugging Face Hub +10 Wauplin, coyotte508, XciD, victor, julien-c, lhoestq, pierric, Sylvestre, hlarcher, rajatarya, seanses, assafvayner • Mar 10 • 194
view reply KV caching enables to re-use what the model previously generated. That way, the model only focuses on the new tokens to generate.Here is an illustrated explanation of KV caching: https://huggingface.co/blog/not-lain/kv-caching
view article Article KV Caching Explained: Optimizing Transformer Inference Efficiency not-lain • Jan 30, 2025 • 325
view article Article Mixture of Experts (MoEs) in Transformers +5 ariG23498, pcuenq, merve, IlyasMoutawwakil, ArthurZ, sergiopaniego, Molbap • Feb 26 • 159
view article Article Compute and Competition in AI: Different FlOPs for Different Folks sasha • Feb 12 • 15
view article Article Custom Kernels for All from Codex and Claude +2 burtenshaw, sayakpaul, ariG23498, evalstate • Feb 13 • 75
view article Article Continuous batching from first principles +1 ror, ArthurZ, mcpotato • Nov 25, 2025 • 378