ce5872705e
### What this PR does / why we need it?
Adds W4A16 quantization method for the Kimi-K2-Thinking model and
updates relevant modules to support the new quantization method.
- Implements complete W4A16 quantization method including weight
packing/unpacking, per-group quantization parameter generation,
post-processing logic and MoE method application.
- Adds parameters `use_int4_w4a16`, `w1_offset` and `w2_offset`, adjusts
`with_quant` conditional logic to support W4A16 matrix multiplication.
- Adds `packed_modules_model_mapping` for Kimi-K2-Thinking model and
processing logic for `weight_packed` field.
- vLLM version: v0.12.0
- vLLM main:
ad32e3e19c
---------
Signed-off-by: zhoux77899 <zhouxiang100@huawei.com>
Signed-off-by: Ruri <33858552+zhoux77899@users.noreply.github.com>
Signed-off-by: Ruri <zhouxiang100@huawei.com>
563 B
563 B
Tutorials
:::{toctree} :caption: Deployment :maxdepth: 1 single_npu single_npu_qwen2.5_vl single_npu_qwen2_audio single_npu_qwen3_embedding single_npu_qwen3_quantization single_npu_qwen3_w4a4 single_node_pd_disaggregation_mooncake multi_npu_qwen3_next multi_npu multi_npu_kimi-k2-thinking multi_npu_moge multi_npu_qwen3_moe multi_npu_quantization single_node_300i DeepSeek-V3.1.md DeepSeek-V3.2-Exp.md Qwen3-235B-A22B.md Qwen3-Coder-30B-A3B multi_node multi_node_kimi multi_node_qwen3vl multi_node_pd_disaggregation_mooncake multi_node_ray Qwen2.5-Omni.md :::