FP8 KV-Cache on Intel® Arc™ Pro B70: 2× Capacity with strong Long Context Throughput Gains
kushal-mittal
• • 3NEW Articles from Team or Enterprise organizations will get promoted to the main section.
Publish, then improve: routing around the 20-Space cap
ParetoOptimal
• Reproducing 147 Machine Learning Papers in 16 Days on One Workstation
ParetoOptimal
• Alpamayo 2 Super: The expert that didn't grow
JonnaMat
• Optimum Intel v2.1.0
echarlaix
• • 5IDEOGRAM-4 for inpainting with Modular Diffusers and Differential Diffusion
OzzyGT
• • 3Waves, queues, and the three free gates
ParetoOptimal
• Picking reproducible papers out of a 6,341-paper haystack
ParetoOptimal
• A $0 fine-tuned judge, and the eval that assigned its seat
ParetoOptimal
• Training a coding agent using the OpenCode harness in remote HF sandboxes with TRL and OpenEnv
sergiopaniego
• • 13Watching a pure-integer GPT think — one degree per frame
nmicic
• iFAN: Training Plain Mask Transformers for the Way They Actually Infer
xxlucas
• • 2AnchorDiT: Semantic Anchor-Guided Transformers for Pixel-Space Image Generation
xxlucas
• • 3The GPU Metrics the Cloud Won't Let You See
nazeerbashashaik
• • 1jina-reranker-v3.5: Faster Listwise Reranking with Hybrid Attention and Self-Distillation
jinaai
• • 5Rivet-1B: A From-Scratch 1.09B Reasoning Model
josh-a
• • 2Build an AI Evaluation from a Hugging Face Dataset Without Writing Python
phranzia
• • 2The Transformers modeling backend serves DeepSeek-R1 at native speed
hmellor
• • 2Language Identification for 42 Indian Languages: Opening New Frontiers in Indic Speech AI
ARTPARK-IISc
• • 6Haawke Verify is now live on Hugging Face Spaces — and it's callable.
inoculatemedia
•