A high-throughput and memory-efficient inference and serving engine for LLMs
Mirrored from https://github.com/tduncan-lum/Nvidia-Optimization-Stack---VLLM.git
Synced Nov 9, 12:30 UTC
Relax protobuf library version constraints (#33202)
Signed-off-by: Jeffrey Wang <jeffreywang@anyscale.com>
Jeffrey Wang committed
Jan 27, 2026 at 20:15 UTC
a97b5e206d78b96a75615f402357dbf7e73d4efe
3 files changed
+6
-6
pyproject.toml
+1
-1
@@ -9,7 +9,7 @@ requires = [
9
"torch == 2.9.1",
10
"wheel",
11
"jinja2",
12
- "grpcio-tools>=1.76.0",
12
+ "grpcio-tools",
13
]
14
build-backend = "setuptools.build_meta"
15
requirements/build.txt
+2
-2
@@ -9,5 +9,5 @@ wheel
9
jinja2>=3.1.6
10
regex
11
build
12
-protobuf>=6.33.2
13
-grpcio-tools>=1.76.0
12
+protobuf
13
+grpcio-tools
requirements/common.txt
+3
-3
@@ -9,7 +9,7 @@ blake3
9
py-cpuinfo
10
transformers >= 4.56.0, < 5
11
tokenizers >= 0.21.1 # Required for fast incremental detokenization.
12
-protobuf >= 6.30.0 # Required by LlamaTokenizer, gRPC.
12
+protobuf # Required by LlamaTokenizer, gRPC.
13
fastapi[standard] >= 0.115.0 # Required by FastAPI's form models in the OpenAI API server's audio transcriptions endpoint.
14
aiohttp
15
openai >= 1.99.1 # For Responses API with reasoning content
@@ -51,5 +51,5 @@ openai-harmony >= 0.0.3 # Required for gpt-oss
51
anthropic >= 0.71.0
52
model-hosting-container-standards >= 0.1.13, < 1.0.0
53
mcp
54
-grpcio>=1.76.0
55
-grpcio-reflection>=1.76.0
\ No newline at end of file
54
+grpcio
55
+grpcio-reflection
\ No newline at end of file