NVIDIA releases official NVFP4 quantization of Qwen 3.8 27B
llm_wizard · x · 2026-09-09
NVIDIA AI confirmed the release of an NVFP4 quantized version of Qwen 3.8 27B, now available on Hugging Face. NVFP4 is NVIDIA's 4-bit floating-point format that shrinks model size and boosts inference efficiency while retaining accuracy, making the Qwen model easier to run locally on RTX and data center GPUs.
Related event: Nvidia Releases NVFP4 Quantized Qwen3.8-27B, Tops Hugging Face Trending(3 posts)→
More from Infra
- Anthropic's ~11 GW contracted compute estimated at ~42% NVIDIA GPUs, ~4.5 GW Google TPU — ns123abc · 2026-10-27
- EdgeRock-4B holds KV cache at 72 MiB across 256K context, 99.1% smaller than Qwen3.5-4B — alifcoder · 2026-10-27
- RTX 5080 price jumps from $1658 to $2318 in ten days on Amazon — Frizzy-MacDrizzle · 2026-10-27
- ComfyUI on 12GB VRAM: measured peaks show offloading turns 24s into 11 minutes — Impossible-Mail-4619 · 2026-10-27
- Neon rebuilds its docs search entirely on its own stack: Lakebase Search, Functions, AI Gateway — matei_zaharia · 2026-10-27
- PyTorch details 2.12x faster on-device streaming ASR on Arm SME2 via ExecuTorch — PyTorch · 2026-10-27