Qwen3.6 27B hits 366 t/s with NVFP4 on V100s

Community post reports 366 tokens/sec (single stream) for Qwen3.6 27B with NVFP4 quantization on V100 GPUs, following an earlier post showing 1000 t/s with the same model and hardware.
1 source
Daily brief
Get tomorrow's AI brief in your inbox
More stories today
- VisionDepth3D creates 3D from 2D videos with AI depth mapping
- xAI SDK v1.18.0 adds grok-4.6 and xhigh reasoning_effort
- WhisperKit enables local speech-to-text transcription on macOS
- Autoware accelerates autonomous vehicle deployment with open-source stack
- Perplexity reportedly offered to acquire Google Chrome one year ago