diff --git a/llm-server/llm/ansible/roles/benchmark/defaults/main.yml b/llm-server/llm/ansible/roles/benchmark/defaults/main.yml index f88ec5f..d912d57 100644 --- a/llm-server/llm/ansible/roles/benchmark/defaults/main.yml +++ b/llm-server/llm/ansible/roles/benchmark/defaults/main.yml @@ -30,7 +30,11 @@ benchmark_task_args: model: dummy-model-name base_url: "{{ benchmark_engine_url }}/chat/completions" num_concurrent: 1 - max_retries: 2 + # xhigh thinking can run to the 32768-token output cap, which takes ~30.2 + # minutes at ~18 t/s; 2700 s covers cap-bound generations plus prompt + # evaluation. Retries are pointless for sequential requests (a timed-out + # generation would just repeat); observed 02.11.2026 when Swift 1.0 + # exceeded the previous 1800 s timeout on AIME. + max_retries: 1 tokenized_requests: false - # xhigh thinking on this hardware needs well over the 300 s default. - timeout: 1800 + timeout: 2700