biondizzle/vllm
1
0
Fork 0
Code Issues Pull Requests Actions 2 Packages Projects Releases Wiki Activity

Re-enable Xet on TPU tests now that hf_xet has been updated (#22666)

Browse Source 
Signed-off-by: Harry Mellor <19981378+hmellor@users.noreply.github.com>
This commit is contained in:
Harry Mellor
2025-08-12 03:54:40 +01:00
committed by GitHub
parent 9b94d6ec8f
commit 839ab00349
3 changed files with 2 additions and 5 deletions
Download Patch File Download Diff File Expand all files Collapse all files

3
tests/entrypoints/llm/test_accuracy.py
View File

@@ -96,9 +96,6 @@ def test_lm_eval_accuracy_v1_engine_fp8_kv_cache(
more_args = None
if current_platform.is_tpu():
# Limit compilation time for TPU V1
# xet doesn't work well for Qwen/Qwen3-1.7B
m.setenv("HF_HUB_DISABLE_XET", "1")
more_args = "max_model_len=2048,max_num_seqs=128,kv_cache_dtype=fp8"
# Add TP test (if provided)