Qwen3.8-27B-W4A16
W4A16 quantization of Qwen/Qwen3.8-27B, produced with llm-compressor via model_free_ptq (data-free).
Vision encoder, gated-delta attention (conv1d / linear_attn), lm_head, embeddings, and norm layers are kept in original precision.
- Downloads last month
- 220
Model tree for lokeshe09/Qwen3.8-27B-W4A16
Base model
Qwen/Qwen3.8-27B