@tduncan / Nvidia-Optimization-Stack-VLLM Mirror · read-only / commits / f8a12ecc7f
Fork 0

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, 04:15 UTC

[Misc] Bump transformers version (#3592)

Roger Wang committed Mar 24, 2024 at 06:32 UTC f8a12ecc7f7ebcffe26e1ae405c6aa533fc400cd
2 files changed +3 -3
requirements-rocm.txt
+2 -2
@@ -7,9 +7,9 @@ ray >= 2.9
7 sentencepiece # Required for LLaMA tokenizer.
8 numpy
9 tokenizers>=0.15.0
10 -transformers >= 4.39.0 # Required for StarCoder2.
10 +transformers >= 4.39.1 # Required for StarCoder2 & Llava.
11 fastapi
12 uvicorn[standard]
13 pydantic >= 2.0 # Required for OpenAI server.
14 prometheus_client >= 0.18.0
15 -outlines == 0.0.34
\ No newline at end of file
15 +outlines == 0.0.34
requirements.txt
+1 -1
@@ -5,7 +5,7 @@ ray >= 2.9
5 sentencepiece # Required for LLaMA tokenizer.
6 numpy
7 torch == 2.1.2
8 -transformers >= 4.39.0 # Required for StarCoder2.
8 +transformers >= 4.39.1 # Required for StarCoder2 & Llava.
9 xformers == 0.0.23.post1 # Required for CUDA 12.1.
10 fastapi
11 uvicorn[standard]