expose flex block size for batch invariant mode (#41252)

Signed-off-by: Angel Li <liangel@meta.com>
This commit is contained in:
liangel-02
2026-05-13 14:11:57 -07:00
committed by GitHub
parent 8efd508204
commit 6b5c389ee3
4 changed files with 114 additions and 8 deletions
+11 -1
View File
@@ -147,8 +147,14 @@ def test_v1_generation_is_deterministic_across_batch_sizes_with_needle(
"backend",
BACKENDS,
)
@pytest.mark.parametrize(
"block_m,block_n",
[(16, 16), (8, 16)],
)
def test_logprobs_bitwise_batch_invariance_bs1_vs_bsN(
backend,
block_m,
block_n,
):
seed = int(os.getenv("VLLM_TEST_SEED", "12345"))
random.seed(seed)
@@ -172,7 +178,11 @@ def test_logprobs_bitwise_batch_invariance_bs1_vs_bsN(
max_model_len=8192,
dtype="auto", # not everything is supported
gpu_memory_utilization=0.9,
attention_config={"backend": backend},
attention_config={
"backend": backend,
"flex_attn_block_m": block_m,
"flex_attn_block_n": block_n,
},
)
# Use more realistic prompts for better token generation