nvidia/Qwen2.5-VL-7B-Instruct-FP8

Quantization vs Qwen/Qwen2.5-VL-7B-Instruct. nvidia/Qwen2.5-VL-7B-Instruct-FP8 is a quantized version of the other (F8_E4M3/BF16).

On the Hub · Downloads on Model Pulse