diff --git a/tests/functional_tests/test_cases/gpt/gpt_grpo_tp4_pp1_dp2_8b_cudagraphs_throughput/model_config.yaml b/tests/functional_tests/test_cases/gpt/gpt_grpo_tp4_pp1_dp2_8b_cudagraphs_throughput/model_config.yaml index f3c0c4ecc5b..b5f735facd5 100644 --- a/tests/functional_tests/test_cases/gpt/gpt_grpo_tp4_pp1_dp2_8b_cudagraphs_throughput/model_config.yaml +++ b/tests/functional_tests/test_cases/gpt/gpt_grpo_tp4_pp1_dp2_8b_cudagraphs_throughput/model_config.yaml @@ -13,6 +13,10 @@ MODEL_ARGS: --inference-dynamic-batching-unified-memory-level: 1 --inference-dynamic-batching-buffer-size-gb: 20 --ckpt-format: torch_dist + # Pin the pre-#3509 (linear) CUDA-graph sizing distribution. #3509 switched the + # default to exponential + a mixed-prefill grid, which raised peak memory and + # tripped this test's mem-allocated-bytes guardrail (~69GB vs ~61GB golden, +13.6%). + --inference-dynamic-batching-cuda-graph-sizing-distribution: linear --seq-length: 1024 --inference-max-seq-length: 1024 --load: ${CHECKPOINT_LOAD_PATH}/model/qwen3-8b-dist diff --git a/tests/functional_tests/test_cases/gpt/gpt_grpo_tp4_pp1_dp2_8b_throughput/model_config.yaml b/tests/functional_tests/test_cases/gpt/gpt_grpo_tp4_pp1_dp2_8b_throughput/model_config.yaml index 80664dcdc59..722c746c103 100644 --- a/tests/functional_tests/test_cases/gpt/gpt_grpo_tp4_pp1_dp2_8b_throughput/model_config.yaml +++ b/tests/functional_tests/test_cases/gpt/gpt_grpo_tp4_pp1_dp2_8b_throughput/model_config.yaml @@ -13,6 +13,10 @@ MODEL_ARGS: --inference-dynamic-batching-unified-memory-level: 1 --inference-dynamic-batching-buffer-size-gb: 20 --ckpt-format: torch_dist + # Pin the pre-#3509 (linear) CUDA-graph sizing distribution. #3509 switched the + # default to exponential + a mixed-prefill grid, which raised peak memory and + # tripped this test's mem-allocated-bytes guardrail (~69GB vs ~61GB golden, +13.6%). + --inference-dynamic-batching-cuda-graph-sizing-distribution: linear --seq-length: 1024 --inference-max-seq-length: 1024 --load: ${CHECKPOINT_LOAD_PATH}/model/qwen3-8b-dist