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, 03:00 UTC
[Misc] Minor fixes in requirements.txt (#3769)
Woosuk Kwon committed
Apr 1, 2024 at 03:15 UTC
f03cc667a09bce92e09365893603f7ec0d87c9f2
2 files changed
+1
-1
requirements-rocm.txt
+1
@@ -15,3 +15,4 @@ uvicorn[standard]
15
pydantic >= 2.0 # Required for OpenAI server.
16
prometheus_client >= 0.18.0
17
outlines == 0.0.34
18
+tiktoken == 0.6.0 # Required for DBRX tokenizer
requirements.txt
-1
@@ -6,7 +6,6 @@ sentencepiece # Required for LLaMA tokenizer.
6
numpy
7
torch == 2.1.2
8
requests
9
-psutil
9
py-cpuinfo
10
transformers >= 4.39.1 # Required for StarCoder2 & Llava.
11
xformers == 0.0.23.post1 # Required for CUDA 12.1.