GLM-4.1V-9B-Thinking-8bit / generation_config.json
axiom-of-choice's picture
Add MLX 8-bit conversion of GLM-4.1V-9B-Thinking with vision preserved and weight-level fidelity vs bf16
9677807 verified
Raw
History Blame Contribute Delete
239 Bytes
{
"_from_model_config": true,
"do_sample": true,
"eos_token_id": [
151329,
151336,
151338,
151348
],
"pad_token_id": 151329,
"top_p": 0.6,
"temperature": 0.8,
"top_k": 2,
"transformers_version": "4.57.1"
}