NVIDIA releases Qwen3.6-27B-NVFP4 quantized model

Hugging Face Models

NVIDIA released Qwen3.6-27B-NVFP4, a 27B-parameter NVFP4 quantized model trending with 47 likes and 58 downloads on HuggingFace. NVFP4 is NVIDIA's 4-bit floating-point format for efficient inference.

NvidiaQwenNVFP4

See 4 more sources

Qwen 3.6 27B is the sweet spot for local development1 day agostaredDiscuss

Qwen 3.6 27B Speculative Decoding Bench: Pushing ~100 TPS on a single RTX 309021 hours agoold-mikeDiscuss

nvidia/Qwen3.6-27B-NVFP4 just dropped1 day agovanbukinDiscuss

Been running Qwen3.6-27B through a 3-critic harness. The harness matters more than I thought1 day agoworkout_JKDiscuss

Hugging Face Models·nvidia

Jun 22, 12:24 PM