CodeStreet
/

Qwen-3.5-9B-FP8-A16

compressed-tensors

Model card Files Files and versions

Qwen-3.5-9B-FP8-A16 / generation_config.json

CodeStreet's picture

Qwen3.5-9B FP8 weight-only (W8A16Fp8): per-channel fp8 weights, bf16 activations

9686523 verified 6 days ago

History Blame Contribute Delete

116 Bytes

	{
	"_from_model_config": true,
	"eos_token_id": 248044,
	"transformers_version": "5.10.1",
	"use_cache": true
	}