Add lm-eval directly to requirements-test.txt (#9161)

2024-10-08 21:22:31 -04:00
parent 2a131965a8
commit 9ba0bd6aa6
5 changed files with 4 additions and 6 deletions
--- a/docs/source/quantization/fp8.rst
+++ b/docs/source/quantization/fp8.rst
@@ -106,7 +106,7 @@ Install ``vllm`` and ``lm-evaluation-harness``:

 .. code-block:: console

-   $ pip install vllm lm_eval==0.4.3
+   $ pip install vllm lm-eval==0.4.4

 Load and run the model in ``vllm``: