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 17, 06:45 UTC
lock torch version to 2.0.1 (#1290)
yanxiyue committed
Oct 11, 2023 at 00:21 UTC
6a6119554cee13bb4ef5018155be5530d6aaca26
2 files changed
+3
-3
pyproject.toml
+1
-1
@@ -3,7 +3,7 @@ requires = [
3
"ninja",
4
"packaging",
5
"setuptools",
6
- "torch >= 2.0.0",
6
+ "torch == 2.0.1",
7
"wheel",
8
]
9
build-backend = "setuptools.build_meta"
requirements.txt
+2
-2
@@ -5,9 +5,9 @@ pandas # Required for Ray data.
5
pyarrow # Required for Ray data.
6
sentencepiece # Required for LLaMA tokenizer.
7
numpy
8
-torch >= 2.0.0
8
+torch == 2.0.1
9
transformers >= 4.33.1 # Required for Code Llama.
10
-xformers >= 0.0.22
10
+xformers == 0.0.22
11
fastapi
12
uvicorn[standard]
13
pydantic < 2 # Required for OpenAI server.