# Mathematical Reasoning on AIME 2024 (Avg@k and Pass@k Evaluation)

Source: [Mathematical Reasoning on AIME 2024 (Avg@k and Pass@k Evaluation)](<https://www.sota2.com/research/sota/mathematical-reasoning-on-aime-2024-avg-at-k-and-pass-at-k-evaluation>)

Collection index: [llms.txt](<https://www.sota2.com/research/llms.txt>)

Data updated: 2026-05-11T10:11:09.490Z

Evaluation of model performance on the American Invitational Mathematics Examination (AIME) 2024 dataset, measured using Average Recall (avg@k) and Pass@k metrics for k=1 and k=8.

A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage.

| Scope | Recorded value |
| --- | --- |
| Task | Mathematical Reasoning |
| Dataset (including any stated split) | AIME competition-level 2024 |
| Metric | Avg@1 |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 16 of 16 |
| Coverage | All results currently returned for this benchmark |

Evaluation settings may differ between results. The original papers and recorded configurations are needed to establish comparability. A shared metric name alone does not establish an identical protocol.

## Metrics

- [Avg@1](<https://www.sota2.com/research/sota/mathematical-reasoning-on-aime-2024-avg-at-k-and-pass-at-k-evaluation.md?metric=Avg%401>): Higher is better; unit: PERCENT.

- [Avg@8](<https://www.sota2.com/research/sota/mathematical-reasoning-on-aime-2024-avg-at-k-and-pass-at-k-evaluation.md?metric=Avg%408>): Higher is better; unit: PERCENT.

- [Pass@1](<https://www.sota2.com/research/sota/mathematical-reasoning-on-aime-2024-avg-at-k-and-pass-at-k-evaluation.md?metric=Pass%401>): Higher is better; unit: PERCENT.

- [Pass@8](<https://www.sota2.com/research/sota/mathematical-reasoning-on-aime-2024-avg-at-k-and-pass-at-k-evaluation.md?metric=Pass%408>): Higher is better; unit: PERCENT.

## Recorded results

| Method | Avg@1 | Result date | Report paper |
| --- | --- | --- | --- |
| GPT-5 nano | 73.3 | Not recorded | 6a0162badbc836b87d82332a |
| Qwen3-8B | 70 | Not recorded | 6a0162badbc836b87d82332a |
| gpt-oss-20b | 70 | Not recorded | 6a0162badbc836b87d82332a |
| Qwen3-4B | 60 | Not recorded | 6a0162badbc836b87d82332a |
| EngGPT2-16B-A3B | 53.3 | Not recorded | 6a0162badbc836b87d82332a |
| Ministral-3-8B | 23.3 | Not recorded | 6a0162badbc836b87d82332a |
| gemma-3-12b-it | 16.7 | Not recorded | 6a0162badbc836b87d82332a |
| gemma-3-4b-it | 13.3 | Not recorded | 6a0162badbc836b87d82332a |
| Llama-3.2-3B-Instruct | 3.3 | Not recorded | 6a0162badbc836b87d82332a |
| Moonlight-16B-A3B-Instruct | 3.3 | Not recorded | 6a0162badbc836b87d82332a |
| FastwebMIIA-7B | 0 | Not recorded | 6a0162badbc836b87d82332a |
| Minerva-7B-instruct-v1.0 | 0 | Not recorded | 6a0162badbc836b87d82332a |
| Velvet-14B | 0 | Not recorded | 6a0162badbc836b87d82332a |
| LLaMAntino-3-ANITA-8B | 0 | Not recorded | 6a0162badbc836b87d82332a |
| Llama-3.1-8B-Instruct | 0 | Not recorded | 6a0162badbc836b87d82332a |
| deepseek-moe-16b-chat | 0 | Not recorded | 6a0162badbc836b87d82332a |

### GPT-5 nano

Avg@1: 73.3. Result ID: 6a01ab3c47f34ce924e41e61.

[Reporting paper](<https://www.sota2.com/research/paper/6a0162badbc836b87d82332a.md>)

Recorded evaluation configuration: {"Model Version":"nano"}. This record does not assert independent reproduction.

### Qwen3-8B

Avg@1: 70. Result ID: 6a01ab3c47f34ce924e41e5b.

[Reporting paper](<https://www.sota2.com/research/paper/6a0162badbc836b87d82332a.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa788e1e209131cea79116.md>)

Recorded evaluation configuration: {"Parameters":"8B"}. This record does not assert independent reproduction.

### gpt-oss-20b

Avg@1: 70. Result ID: 6a01ab3c47f34ce924e41e63.

[Reporting paper](<https://www.sota2.com/research/paper/6a0162badbc836b87d82332a.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7ab01e209131cea84a0c.md>)

Recorded evaluation configuration: {"Parameters":"20b"}. This record does not assert independent reproduction.

### Qwen3-4B

Avg@1: 60. Result ID: 6a01ab3c47f34ce924e41e5a.

[Reporting paper](<https://www.sota2.com/research/paper/6a0162badbc836b87d82332a.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa788e1e209131cea79116.md>)

Recorded evaluation configuration: {"Parameters":"4B"}. This record does not assert independent reproduction.

### EngGPT2-16B-A3B

Avg@1: 53.3. Result ID: 6a01ab3c47f34ce924e41e55.

[Reporting paper](<https://www.sota2.com/research/paper/6a0162badbc836b87d82332a.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6a0162badbc836b87d82332a.md>)

Recorded evaluation configuration: {"Parameters":"16B"}. This record does not assert independent reproduction.

### Ministral-3-8B

Avg@1: 23.3. Result ID: 6a01ab3c47f34ce924e41e60.

[Reporting paper](<https://www.sota2.com/research/paper/6a0162badbc836b87d82332a.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69672325e25ad2fa9d1ec218.md>)

Recorded evaluation configuration: {"Parameters":"8B"}. This record does not assert independent reproduction.

### gemma-3-12b-it

Avg@1: 16.7. Result ID: 6a01ab3c47f34ce924e41e5f.

[Reporting paper](<https://www.sota2.com/research/paper/6a0162badbc836b87d82332a.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa779f1e209131cea746cd.md>)

Recorded evaluation configuration: {"Parameters":"12b"}. This record does not assert independent reproduction.

### gemma-3-4b-it

Avg@1: 13.3. Result ID: 6a01ab3c47f34ce924e41e5e.

[Reporting paper](<https://www.sota2.com/research/paper/6a0162badbc836b87d82332a.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa779f1e209131cea746cd.md>)

Recorded evaluation configuration: {"Parameters":"4b"}. This record does not assert independent reproduction.

### Llama-3.2-3B-Instruct

Avg@1: 3.3. Result ID: 6a01ab3c47f34ce924e41e5c.

[Reporting paper](<https://www.sota2.com/research/paper/6a0162badbc836b87d82332a.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa75d11e209131cea6a028.md>)

[Implementation](<https://github.com/meta-llama/llama-agentic-system>)

Recorded evaluation configuration: {"Parameters":"3B"}. This record does not assert independent reproduction.

### Moonlight-16B-A3B-Instruct

Avg@1: 3.3. Result ID: 6a01ab3c47f34ce924e41e64.

[Reporting paper](<https://www.sota2.com/research/paper/6a0162badbc836b87d82332a.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa771a1e209131cea70fcd.md>)

Recorded evaluation configuration: {"Parameters":"16B"}. This record does not assert independent reproduction.

### FastwebMIIA-7B

Avg@1: 0. Result ID: 6a01ab3c47f34ce924e41e56.

[Reporting paper](<https://www.sota2.com/research/paper/6a0162badbc836b87d82332a.md>)

Recorded evaluation configuration: {"Parameters":"7B"}. This record does not assert independent reproduction.

### Minerva-7B-instruct-v1.0

Avg@1: 0. Result ID: 6a01ab3c47f34ce924e41e57.

[Reporting paper](<https://www.sota2.com/research/paper/6a0162badbc836b87d82332a.md>)

Recorded evaluation configuration: {"Parameters":"7B"}. This record does not assert independent reproduction.

### Velvet-14B

Avg@1: 0. Result ID: 6a01ab3c47f34ce924e41e58.

[Reporting paper](<https://www.sota2.com/research/paper/6a0162badbc836b87d82332a.md>)

Recorded evaluation configuration: {"Parameters":"14B"}. This record does not assert independent reproduction.

### LLaMAntino-3-ANITA-8B

Avg@1: 0. Result ID: 6a01ab3c47f34ce924e41e59.

[Reporting paper](<https://www.sota2.com/research/paper/6a0162badbc836b87d82332a.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa73321e209131cea5b137.md>)

Recorded evaluation configuration: {"Parameters":"8B"}. This record does not assert independent reproduction.

### Llama-3.1-8B-Instruct

Avg@1: 0. Result ID: 6a01ab3c47f34ce924e41e5d.

[Reporting paper](<https://www.sota2.com/research/paper/6a0162badbc836b87d82332a.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa75d11e209131cea6a028.md>)

[Implementation](<https://github.com/meta-llama/llama-agentic-system>)

Recorded evaluation configuration: {"Parameters":"8B"}. This record does not assert independent reproduction.

### deepseek-moe-16b-chat

Avg@1: 0. Result ID: 6a01ab3c47f34ce924e41e62.

[Reporting paper](<https://www.sota2.com/research/paper/6a0162badbc836b87d82332a.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa71c71e209131cea52e9d.md>)

[Implementation](<https://github.com/deepseek-ai/DeepSeek-MoE>)

Recorded evaluation configuration: {"Parameters":"16b"}. This record does not assert independent reproduction.

Page 1 of 1. This is an index page, not the full collection.
