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/examples/offline_inference.py
+++ b/server/vllm/examples/offline_inference.py
--- a/server/vllm/examples/openai_chatcompletion_client.py
+++ b/server/vllm/examples/openai_chatcompletion_client.py
--- a/server/vllm/examples/openai_completion_client.py
+++ b/server/vllm/examples/openai_completion_client.py
--- a/server/vllm/format.sh
+++ b/server/vllm/format.sh
--- a/server/vllm/mypy.ini
+++ b/server/vllm/mypy.ini
+[mypy]
+python_version = 3.8
+ignore_missing_imports = True
+files = vllm
+# TODO(woosuk): Include the code from Megatron and HuggingFace.
+exclude = vllm/model_executor/parallel_utils/|vllm/model_executor/models/
--- a/server/vllm/pyproject.toml
+++ b/server/vllm/pyproject.toml
--- a/server/vllm/requirements-dev.txt
+++ b/server/vllm/requirements-dev.txt
--- a/server/vllm/requirements.txt
+++ b/server/vllm/requirements.txt
--- a/server/vllm/setup.py
+++ b/server/vllm/setup.py
--- a/server/vllm/tests/async_engine/api_server_async_engine.py
+++ b/server/vllm/tests/async_engine/api_server_async_engine.py
--- a/server/vllm/tests/async_engine/test_api_server.py
+++ b/server/vllm/tests/async_engine/test_api_server.py
--- a/server/vllm/tests/async_engine/test_async_llm_engine.py
+++ b/server/vllm/tests/async_engine/test_async_llm_engine.py
--- a/server/vllm/tests/async_engine/test_request_tracker.py
+++ b/server/vllm/tests/async_engine/test_request_tracker.py
--- a/server/vllm/tests/conftest.py
+++ b/server/vllm/tests/conftest.py
--- a/server/vllm/tests/distributed/test_comm_ops.py
+++ b/server/vllm/tests/distributed/test_comm_ops.py
--- a/server/vllm/tests/engine/test_detokenize.py
+++ b/server/vllm/tests/engine/test_detokenize.py
--- a/server/vllm/tests/kernels/conftest.py
+++ b/server/vllm/tests/kernels/conftest.py
--- a/server/vllm/tests/kernels/test_activation.py
+++ b/server/vllm/tests/kernels/test_activation.py
--- a/server/vllm/tests/kernels/test_attention.py
+++ b/server/vllm/tests/kernels/test_attention.py
--- a/server/vllm/tests/kernels/test_cache.py
+++ b/server/vllm/tests/kernels/test_cache.py