diff --git a/tests/evals/gsm8k/configs/DeepSeek-R1-DP.yaml b/tests/evals/gsm8k/configs/DeepSeek-R1-DP.yaml index f351a17220..0c6a598a8a 100644 --- a/tests/evals/gsm8k/configs/DeepSeek-R1-DP.yaml +++ b/tests/evals/gsm8k/configs/DeepSeek-R1-DP.yaml @@ -8,4 +8,4 @@ server_args: >- --max-model-len 4096 --data-parallel-size 8 --enable-expert-parallel - --speculative-config '{"method":"mtp","num_speculative_tokens":1}' + --speculative-config '{"method":"mtp","num_speculative_tokens":3}' diff --git a/tests/evals/gsm8k/configs/DeepSeek-R1-TP.yaml b/tests/evals/gsm8k/configs/DeepSeek-R1-TP.yaml index ba3463463b..f6ab810085 100644 --- a/tests/evals/gsm8k/configs/DeepSeek-R1-TP.yaml +++ b/tests/evals/gsm8k/configs/DeepSeek-R1-TP.yaml @@ -8,4 +8,4 @@ server_args: >- --max-model-len 4096 --tensor-parallel-size 8 --enable-expert-parallel - --speculative-config '{"method":"mtp","num_speculative_tokens":1}' + --speculative-config '{"method":"mtp","num_speculative_tokens":3}' diff --git a/tests/evals/gsm8k/configs/DeepSeek-V3.2-DP.yaml b/tests/evals/gsm8k/configs/DeepSeek-V3.2-DP.yaml index d7d1df974a..c0e2e8f044 100644 --- a/tests/evals/gsm8k/configs/DeepSeek-V3.2-DP.yaml +++ b/tests/evals/gsm8k/configs/DeepSeek-V3.2-DP.yaml @@ -8,4 +8,4 @@ server_args: >- --max-model-len 4096 --data-parallel-size 8 --enable-expert-parallel - --speculative-config '{"method":"mtp","num_speculative_tokens":1}' + --speculative-config '{"method":"mtp","num_speculative_tokens":3}' diff --git a/tests/evals/gsm8k/configs/DeepSeek-V3.2-TP.yaml b/tests/evals/gsm8k/configs/DeepSeek-V3.2-TP.yaml index 83687594d4..d31c63b8d7 100644 --- a/tests/evals/gsm8k/configs/DeepSeek-V3.2-TP.yaml +++ b/tests/evals/gsm8k/configs/DeepSeek-V3.2-TP.yaml @@ -8,4 +8,4 @@ server_args: >- --max-model-len 4096 --tensor-parallel-size 8 --enable-expert-parallel - --speculative-config '{"method":"mtp","num_speculative_tokens":1}' + --speculative-config '{"method":"mtp","num_speculative_tokens":3}'