Hcompany/Holo-3.1-35B-A3B-FP8

Quantization vs Qwen/Qwen3.5-0.8B. Hcompany/Holo-3.1-35B-A3B-FP8 is a quantized version of the other (F8_E4M3/BF16).

On the Hub · Downloads on Model Pulse