Merge remote-tracking branch 'origin/v0.7.2-dev' into v0.7.2_zero_overhead

ca4ec0ce · lizhigong · 0be169ad · ae0ed592 · ca4ec0ce · ca4ec0ce
Commit ca4ec0ce authored Mar 25, 2025 by lizhigong
10 changed files
--- a/vllm/model_executor/layers/quantization/utils/int8_utils.py
+++ b/vllm/model_executor/layers/quantization/utils/int8_utils.py
--- a/vllm/model_executor/model_loader/utils.py
+++ b/vllm/model_executor/model_loader/utils.py
--- a/vllm/model_executor/models/deepseek_v2.py
+++ b/vllm/model_executor/models/deepseek_v2.py
--- a/vllm/model_executor/models/gpt_neox.py
+++ b/vllm/model_executor/models/gpt_neox.py
--- a/vllm/model_executor/models/llama.py
+++ b/vllm/model_executor/models/llama.py
--- a/vllm/model_executor/models/qwen.py
+++ b/vllm/model_executor/models/qwen.py
--- a/vllm/model_executor/models/qwen2.py
+++ b/vllm/model_executor/models/qwen2.py
--- a/vllm/platforms/rocm.py
+++ b/vllm/platforms/rocm.py
--- a/vllm/spec_decode/spec_decode_worker.py
+++ b/vllm/spec_decode/spec_decode_worker.py
--- a/vllm/utils.py
+++ b/vllm/utils.py