[misc] Use --cuda-graph-max-bs-decode in tests, examples, and docs (#29591)

2026-06-28 18:38:28 -07:00
parent 3217410cf6
commit 909123ddb8
140 changed files with 304 additions and 290 deletions
@@ -21,7 +21,7 @@ def main():
        speculative_num_steps=3,
        speculative_eagle_topk=4,
        speculative_num_draft_tokens=16,
-        cuda_graph_max_bs=8,
+        cuda_graph_max_bs_decode=8,
    )

    outputs = llm.generate(prompts, sampling_params)