add custom vllm source code

70056d1e · huangwb · 12d93ad7 · 70056d1e · 70056d1e · 70056d1e
Commit 70056d1e authored May 29, 2024 by huangwb
20 changed files
--- a/server/vllm/vllm/model_executor/models/bloom.py
+++ b/server/vllm/vllm/model_executor/models/bloom.py
--- a/server/vllm/vllm/model_executor/models/falcon.py
+++ b/server/vllm/vllm/model_executor/models/falcon.py
--- a/server/vllm/vllm/model_executor/models/gpt2.py
+++ b/server/vllm/vllm/model_executor/models/gpt2.py
--- a/server/vllm/vllm/model_executor/models/gpt_bigcode.py
+++ b/server/vllm/vllm/model_executor/models/gpt_bigcode.py
--- a/server/vllm/vllm/model_executor/models/gpt_j.py
+++ b/server/vllm/vllm/model_executor/models/gpt_j.py
--- a/server/vllm/vllm/model_executor/models/gpt_neox.py
+++ b/server/vllm/vllm/model_executor/models/gpt_neox.py
--- a/server/vllm/vllm/model_executor/models/internlm.py
+++ b/server/vllm/vllm/model_executor/models/internlm.py
--- a/server/vllm/vllm/model_executor/models/llama.py
+++ b/server/vllm/vllm/model_executor/models/llama.py
--- a/server/vllm/vllm/model_executor/models/mpt.py
+++ b/server/vllm/vllm/model_executor/models/mpt.py
--- a/server/vllm/vllm/model_executor/models/opt.py
+++ b/server/vllm/vllm/model_executor/models/opt.py
--- a/server/vllm/vllm/model_executor/models/qwen.py
+++ b/server/vllm/vllm/model_executor/models/qwen.py
--- a/server/vllm/vllm/model_executor/parallel_utils/README.md
+++ b/server/vllm/vllm/model_executor/parallel_utils/README.md
--- a/server/vllm/vllm/model_executor/parallel_utils/__init__.py
+++ b/server/vllm/vllm/model_executor/parallel_utils/__init__.py
--- a/server/vllm/vllm/model_executor/parallel_utils/communication_op.py
+++ b/server/vllm/vllm/model_executor/parallel_utils/communication_op.py
--- a/server/vllm/vllm/model_executor/parallel_utils/layers.py
+++ b/server/vllm/vllm/model_executor/parallel_utils/layers.py
--- a/server/vllm/vllm/model_executor/parallel_utils/parallel_state.py
+++ b/server/vllm/vllm/model_executor/parallel_utils/parallel_state.py
--- a/server/vllm/vllm/model_executor/parallel_utils/utils.py
+++ b/server/vllm/vllm/model_executor/parallel_utils/utils.py
--- a/server/vllm/vllm/model_executor/quantization_utils/__init__.py
+++ b/server/vllm/vllm/model_executor/quantization_utils/__init__.py
--- a/server/vllm/vllm/model_executor/quantization_utils/awq.py
+++ b/server/vllm/vllm/model_executor/quantization_utils/awq.py
--- a/server/vllm/vllm/model_executor/quantization_utils/base.py
+++ b/server/vllm/vllm/model_executor/quantization_utils/base.py