| CoTMode | 2B vs 1.5B | 4B vs 2.9B | 9B vs 7.2B | 13.3B | |||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Qwen3.5 | RWKV | Δ | Qwen3.5 | RWKV | Δ | Qwen3.5 | RWKV | Δ | Qwen3.5 | RWKV | Δ | ||||
| agieval | 5151 | avg@1 | NoCoT | — | — | — | — | — | — | — | — | ||||
| aime24 | 30 | avg@128 | CoT | — | — | — | — | — | — | — | — | ||||
| aime25 | 30 | avg@128 | CoT | — | — | — | — | — | — | — | — | ||||
| aimo_progress_prize_1 | 10 | avg@512 | CoT | — | — | — | — | — | — | — | — | ||||
| arc:challenge | 1172 | avg@1 | NoCoT | — | — | — | — | — | — | — | — | ||||
| arc:easy | 2376 | avg@1 | NoCoT | — | — | — | — | — | — | — | — | ||||
| arithmetic | 20000 | avg@1 | FakeCoT | — | — | — | — | — | — | — | — | ||||
| asdiv | 2305 | avg@1 | FakeCoT | — | — | — | — | — | — | — | — | ||||
| bfcl_v4 | 3641 | avg@1 | FakeCoT | — | — | — | — | — | — | — | — | ||||
| bigbench_hard | 13054 | avg@1 | NoCoT | — | — | — | — | — | — | — | — | — | |||
| BigCodeBench | 1140 | avg@4 | NoCoT | — | — | — | — | — | — | — | — | — | |||
| BigCodeBench | 1140 | avg@4 | FakeCoT | — | — | — | — | — | — | — | — | ||||
| ceval_zho_mcf | 1343 | avg@1 | NoCoT | — | — | — | — | — | — | — | — | ||||
| commonsenseqa | 1221 | avg@1 | NoCoT | — | — | — | — | — | — | — | — | ||||
| gaia | 165 | avg@1 | FakeCoT | — | — | — | — | — | — | — | — | ||||
| gpqa:diamond | 198 | avg@1 | CoT | — | — | — | — | — | — | — | — | ||||
| gpqa:main | 448 | avg@1 | CoT | — | — | — | — | — | — | — | — | ||||
| gsm8k | 1319 | avg@1 | CoT | — | — | — | — | — | — | — | — | ||||
| gsm_plus | 9233 | avg@1 | CoT | — | — | — | — | — | — | — | — | ||||
| hellaswag | 10042 | avg@1 | NoCoT | — | — | — | — | — | — | — | — | ||||
| HumanEval | 164 | avg@16 | NoCoT | — | — | — | — | — | — | — | — | — | |||
| HumanEval | 164 | avg@16 | FakeCoT | — | — | — | — | — | — | — | — | ||||
| HumanEval | 164 | avg@16 | CoT | — | — | — | — | — | — | — | — | ||||
| HumanEval+ | 164 | avg@16 | NoCoT | — | — | — | — | — | — | — | — | — | |||
| HumanEval+ | 164 | avg@16 | FakeCoT | — | — | — | — | — | — | — | — | ||||
| HumanEval+ | 164 | avg@16 | CoT | — | — | — | — | — | — | — | — | ||||
| ifbench_multiturn | 1387 | avg@1 | CoT | — | — | — | — | — | — | — | — | ||||
| ifbench_test | 300 | avg@1 | CoT | — | — | — | — | — | — | — | — | ||||
| ifeval | 541 | avg@1 | CoT | — | — | — | — | — | — | — | — | ||||
| lcb:codegeneration | 268 | avg@1 | CoT | — | — | — | — | — | — | — | — | ||||
| math | 5000 | avg@4 | CoT | — | — | — | — | — | — | — | — | ||||
| math_500 | 500 | avg@8 | CoT | — | — | — | — | — | — | — | — | ||||
| mathqa | 2985 | avg@1 | NoCoT | — | — | — | — | — | — | — | — | ||||
| MBPP | 500 | avg@8 | NoCoT | — | — | — | — | — | — | — | — | — | |||
| MBPP | 500 | avg@8 | FakeCoT | — | — | — | — | — | — | — | — | ||||
| MBPP | 500 | avg@8 | CoT | — | — | — | — | — | — | — | — | ||||
| MBPP+ | 378 | avg@8 | NoCoT | — | — | — | — | — | — | — | — | — | |||
| MBPP+ | 378 | avg@8 | FakeCoT | — | — | — | — | — | — | — | — | ||||
| MBPP+ | 378 | avg@8 | CoT | — | — | — | — | — | — | — | — | ||||
| mc_taco | 9442 | avg@1 | CoT | — | — | — | — | — | — | — | — | ||||
| med_mcqa | 4183 | avg@1 | NoCoT | — | — | — | — | — | — | — | — | ||||
| med_qa | 2545 | avg@1 | NoCoT | — | — | — | — | — | — | — | — | ||||
| mmlu | 14042 | avg@1 | NoCoT | — | — | — | — | — | — | — | — | ||||
| mmlu_pro | 12032 | avg@1 | NoCoT | — | — | — | — | — | — | — | — | ||||
| mmlu_redux_2 | 5700 | avg@1 | NoCoT | — | — | — | — | — | — | — | — | ||||
| MultiPL-E HumanEval | 2864 | avg@1 | NoCoT | — | — | — | — | — | — | — | — | — | |||
| MultiPL-E HumanEval | 2864 | avg@1 | FakeCoT | — | — | — | — | — | — | — | — | — | — | ||
| MultiPL-E MBPP | 6987 | avg@1 | NoCoT | — | — | — | — | — | — | — | — | — | |||
| mutual | 886 | avg@4 | CoT | — | — | — | — | — | — | — | — | ||||
| olympiad_bench | 2673 | avg@1 | CoT | — | — | — | — | — | — | — | — | ||||
| openbookqa | 1000 | avg@1 | NoCoT | — | — | — | — | — | — | — | — | ||||
| race | 1045 | avg@4 | CoT | — | — | — | — | — | — | — | — | ||||
| tau2_bench | 278 | avg@1 | NoCoT | — | — | — | — | — | — | — | — | — | — | ||
| tau2_bench | 278 | avg@1 | FakeCoT | — | — | — | — | — | — | — | — | ||||
| tau3_bench | 375 | avg@1 | NoCoT | — | — | — | — | — | — | — | — | — | — | — | |
| tau3_bench | 375 | avg@1 | FakeCoT | — | — | — | — | — | — | — | — | ||||
| tau_bench | 165 | avg@1 | NoCoT | — | — | — | — | — | — | — | — | — | |||
| tau_bench | 165 | avg@1 | FakeCoT | — | — | — | — | — | — | — | — | ||||
| truthfulqa:mc | 817 | avg@1 | NoCoT | — | — | — | — | — | — | — | — | ||||
| winogender | 960 | avg@2 | CoT | — | — | — | — | — | — | — | — | ||||
| winogrande | 1267 | avg@1 | NoCoT | — | — | — | — | — | — | — | — | ||||
| wmdp | 3668 | avg@1 | CoT | — | — | — | — | — | — | — | — | ||||