The typical standard errors between pairs of models on this dataset as a function of the absolute accuracy.
The standard error of each model pair against their observed accuracy difference. Pairs below a reference line differ by more than the corresponding number of standard errors, i.e. they are statistically distinguishable at that level.
| model | pass1 | pass@count | win_rate | count | SE(A) | SE_x(A) | SE_pred(A) |
|---|---|---|---|---|---|---|---|
| google_gemma_3_27b_it | 40.2 | 52.8 | 38.8 | 3 | 3.7 | 3 | 2.2 |
| google_gemma_2_27b_it | 22.2 | 32.8 | 21.1 | 2 | 3.1 | 1.9 | 2.4 |
| google_gemma_3_4b_it | 18 | 28.9 | 17 | 5 | 2.9 | 2.3 | 1.6 |
| google_gemma_2_9b_it | 15 | 28.3 | 14.1 | 4 | 2.7 | 1.8 | 2 |
| llama-3.2-3B-instruct | 8.94 | 9.44 | 8.24 | 10 | 2.1 | 2.1 | 0.51 |
| llama-3.1-8B-instruct | 7.78 | 8.33 | 7.09 | 7 | 2 | 1.9 | 0.5 |
| google_gemma_3_12b_it | 7.5 | 15 | 7.09 | 4 | 2 | 1.4 | 1.4 |
| google_codegemma_1.1_7b_it | 5 | 17.2 | 4.62 | 5 | 1.6 | 0.71 | 1.5 |
| google_gemma_7b_it | 2.92 | 6.11 | 2.64 | 4 | 1.3 | 0.82 | 0.95 |
| llama-3.2-1B-instruct | 2.65 | 2.78 | 2.48 | 13 | 1.2 | 1.2 | 0.24 |
| mistralai_mixtral_8x22b_instruct_v0.1 | 2.59 | 6.67 | 2.4 | 3 | 1.2 | 0.41 | 1.1 |
| google_gemma_3_1b_it | 0.694 | 2.22 | 0.636 | 4 | 0.62 | 0.22 | 0.58 |
| qwen2.5-coder-32b-instruct | 0.278 | 0.556 | 0.278 | 2 | 0.39 | 0 | 0.39 |
| qwen3-32b | 0.185 | 0.556 | 0.176 | 3 | 0.32 | 0 | 0.32 |
| google_gemma_2b_it | 0.139 | 0.556 | 0.129 | 4 | 0.28 | 0 | 0.28 |
| deepseek_r1_distill_qwen_14b | 0 | 0 | 0 | 4 | 0 | 0 | 0 |
| deepseek_r1_distill_qwen_1.5b | 0 | 0 | 0 | 4 | 0 | 0 | 0 |
| deepseek_r1_distill_llama_70b | 0 | 0 | 0 | 2 | 0 | 0 | 0 |
| deepseek_r1_distill_llama_8b | 0 | 0 | 0 | 4 | 0 | 0 | 0 |
| deepseek_v2_lite_chat | 0 | 0 | 0 | 3 | 0 | 0 | 0 |
| deepseek_r1_distill_qwen_7b | 0 | 0 | 0 | 4 | 0 | 0 | 0 |
| deepseek_r1_distill_qwen_32b | 0 | 0 | 0 | 2 | 0 | 0 | 0 |
| mistralai_mathstral_7b_v0.1 | 0 | 0 | 0 | 4 | 0 | 0 | 0 |
| mistralai_mistral_7b_instruct_v0.2 | 0 | 0 | 0 | 4 | 0 | 0 | 0 |
| mistralai_mistral_7b_instruct_v0.3 | 0 | 0 | 0 | 4 | 0 | 0 | 0 |
| mistralai_mistral_7b_instruct_v0.1 | 0 | 0 | 0 | 4 | 0 | 0 | 0 |
| mistralai_ministral_8b_instruct_2410 | 0 | 0 | 0 | 4 | 0 | 0 | 0 |
| qwen1.5-1.8b-chat | 0 | 0 | 0 | 3 | 0 | 0 | 0 |
| qwen1.5-14b-chat | 0 | 0 | 0 | 3 | 0 | 0 | 0 |
| qwen1.5-32b-chat | 0 | 0 | 0 | 3 | 0 | 0 | 0 |
| qwen1.5-72b-chat | 0 | 0 | 0 | 2 | 0 | 0 | 0 |
| qwen1.5-7b-chat | 0 | 0 | 0 | 3 | 0 | 0 | 0 |
| qwen2-0.5b-instruct | 0 | 0 | 0 | 5 | 0 | 0 | 0 |
| mistralai_mixtral_8x7b_instruct_v0.1 | 0 | 0 | 0 | 3 | 0 | 0 | 0 |
| qwen1.5-0.5b-chat | 0 | 0 | 0 | 5 | 0 | 0 | 0 |
| qwen2-72b-instruct | 0 | 0 | 0 | 2 | 0 | 0 | 0 |
| qwen2-1.5b-instruct | 0 | 0 | 0 | 4 | 0 | 0 | 0 |
| qwen2-math-72b-instruct | 0 | 0 | 0 | 2 | 0 | 0 | 0 |
| qwen2-7b-instruct | 0 | 0 | 0 | 4 | 0 | 0 | 0 |
| qwen2-math-7b-instruct | 0 | 0 | 0 | 2 | 0 | 0 | 0 |
| qwen2.5-coder-0.5b-instruct | 0 | 0 | 0 | 5 | 0 | 0 | 0 |
| qwen2.5-coder-1.5b-instruct | 0 | 0 | 0 | 4 | 0 | 0 | 0 |
| qwen2-math-1.5b-instruct | 0 | 0 | 0 | 3 | 0 | 0 | 0 |
| qwen2.5-coder-14b-instruct | 0 | 0 | 0 | 3 | 0 | 0 | 0 |
| qwen2.5-coder-3b-instruct | 0 | 0 | 0 | 4 | 0 | 0 | 0 |
| qwen3-0.6b | 0 | 0 | 0 | 5 | 0 | 0 | 0 |
| qwen2.5-coder-7b-instruct | 0 | 0 | 0 | 4 | 0 | 0 | 0 |
| qwen3-1.7b | 0 | 0 | 0 | 4 | 0 | 0 | 0 |
| qwen3-14b | 0 | 0 | 0 | 3 | 0 | 0 | 0 |
| qwen3-4b | 0 | 0 | 0 | 4 | 0 | 0 | 0 |
| qwen3-8b | 0 | 0 | 0 | 3 | 0 | 0 | 0 |