NVIDIA releases FP4-quantized DeepSeek-V4.1-Flash vision-language model on HF
nvidia · hf · 2026-10-31
NVIDIA published nvidia/DeepSeek-V4.1-Flash-1017-NVFP4 on Hugging Face: an FP4-quantized image-text-to-text model based on the deepseekv41 architecture, produced with the Model Optimizer (ModelOpt) toolchain and shipped as safetensors, targeting efficient low-precision inference deployment.
Related event: NVIDIA releases NVFP4 quantized DeepSeek-V4.1 Flash model(2 posts)→
More from Infra
- Sam Altman calls local inference "a terrible allocation of scarce resources," community pushes back — ngxson · 2026-10-31
- Cerebras breaks down how it deploys multi-trillion parameter models on wafer-scale chips — pritisinghhhh · 2026-10-31
- Sequoia's SpaceX model implies massive 2028 compute demand—but does that much AI demand even exist? — aronchick · 2026-10-31
- Intuit risk-scores 700K assets on Neo4j with 10-12 hop queries in milliseconds — JeremyCMorgan · 2026-10-31
- AWS publishes a five-stage playbook for managing Amazon Bedrock costs end to end — AWS ML Blog · 2026-10-31
- New Yorker: The global chip supply chain is humanity's greatest collective achievement — anselm · 2026-10-31