# Mathematical Reasoning on AMC (pass@1, pass@16)

Source: [Mathematical Reasoning on AMC (pass@1, pass@16)](<https://www.sota2.com/research/sota/mathematical-reasoning-on-amc-pass-at-1-pass-at-16>)

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

Data updated: 2026-02-26T06:11:15.631Z

Evaluation of model performance on the AMC dataset for mathematical reasoning tasks, specifically measured using the pass@1 and pass@16 metrics.

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) | AMC |
| Metric | pass@1 |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 14 of 14 |
| 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

- [pass@1](<https://www.sota2.com/research/sota/mathematical-reasoning-on-amc-pass-at-1-pass-at-16.md?metric=pass%401>): Higher is better; unit: PERCENT.

- [pass@16](<https://www.sota2.com/research/sota/mathematical-reasoning-on-amc-pass-at-1-pass-at-16.md?metric=pass%4016>): Higher is better; unit: PERCENT.

## Recorded results

| Method | pass@1 | Result date | Report paper |
| --- | --- | --- | --- |
| EVOL-RL | 58.8 | Not recorded | 68fa7bc11e209131cea8a32e |
| EVOL-RL | 55.5 | Not recorded | 68fa7bc11e209131cea8a32e |
| EVOL-RL | 54.4 | Not recorded | 68fa7bc11e209131cea8a32e |
| TTRL | 53.6 | Not recorded | 68fa7bc11e209131cea8a32e |
| EVOL-RL | 51.4 | Not recorded | 68fa7bc11e209131cea8a32e |
| TTRL | 51.1 | Not recorded | 68fa7bc11e209131cea8a32e |
| EVOL-RL | 50.3 | Not recorded | 68fa7bc11e209131cea8a32e |
| EVOL-RL | 49.9 | Not recorded | 68fa7bc11e209131cea8a32e |
| TTRL | 49.5 | Not recorded | 68fa7bc11e209131cea8a32e |
| TTRL | 47.6 | Not recorded | 68fa7bc11e209131cea8a32e |
| TTRL | 43.6 | Not recorded | 68fa7bc11e209131cea8a32e |
| TTRL | 42.5 | Not recorded | 68fa7bc11e209131cea8a32e |
| Qwen3-4B-Base | 39.3 | Not recorded | 68fa7bc11e209131cea8a32e |
| Qwen3-8B-Base | 38.7 | Not recorded | 68fa7bc11e209131cea8a32e |

### EVOL-RL

pass@1: 58.8. Result ID: 699ae4ac5698665ec474ca39.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7bc11e209131cea8a32e.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7bc11e209131cea8a32e.md>)

Recorded evaluation configuration: {"Base Model Size":"8B","Training Dataset":"MATH-500"}. This record does not assert independent reproduction.

### EVOL-RL

pass@1: 55.5. Result ID: 699ae4ac5698665ec474ca37.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7bc11e209131cea8a32e.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7bc11e209131cea8a32e.md>)

Recorded evaluation configuration: {"Base Model Size":"8B","Training Dataset":"MATH-TRAIN"}. This record does not assert independent reproduction.

### EVOL-RL

pass@1: 54.4. Result ID: 699ae4ac5698665ec474ca3b.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7bc11e209131cea8a32e.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7bc11e209131cea8a32e.md>)

Recorded evaluation configuration: {"Base Model Size":"8B","Training Dataset":"AIME24"}. This record does not assert independent reproduction.

### TTRL

pass@1: 53.6. Result ID: 699ae4ac5698665ec474ca36.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7bc11e209131cea8a32e.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa79ce1e209131cea7f7aa.md>)

Recorded evaluation configuration: {"Base Model Size":"8B","Training Dataset":"MATH-TRAIN"}. This record does not assert independent reproduction.

### EVOL-RL

pass@1: 51.4. Result ID: 699ae4ac5698665ec474ca30.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7bc11e209131cea8a32e.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7bc11e209131cea8a32e.md>)

Recorded evaluation configuration: {"Base Model Size":"4B","Training Dataset":"MATH-TRAIN"}. This record does not assert independent reproduction.

### TTRL

pass@1: 51.1. Result ID: 699ae4ac5698665ec474ca38.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7bc11e209131cea8a32e.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa79ce1e209131cea7f7aa.md>)

Recorded evaluation configuration: {"Base Model Size":"8B","Training Dataset":"MATH-500"}. This record does not assert independent reproduction.

### EVOL-RL

pass@1: 50.3. Result ID: 699ae4ac5698665ec474ca32.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7bc11e209131cea8a32e.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7bc11e209131cea8a32e.md>)

Recorded evaluation configuration: {"Base Model Size":"4B","Training Dataset":"MATH-500"}. This record does not assert independent reproduction.

### EVOL-RL

pass@1: 49.9. Result ID: 699ae4ac5698665ec474ca34.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7bc11e209131cea8a32e.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7bc11e209131cea8a32e.md>)

Recorded evaluation configuration: {"Base Model Size":"4B","Training Dataset":"AIME24"}. This record does not assert independent reproduction.

### TTRL

pass@1: 49.5. Result ID: 699ae4ac5698665ec474ca3a.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7bc11e209131cea8a32e.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa79ce1e209131cea7f7aa.md>)

Recorded evaluation configuration: {"Base Model Size":"8B","Training Dataset":"AIME24"}. This record does not assert independent reproduction.

### TTRL

pass@1: 47.6. Result ID: 699ae4ac5698665ec474ca31.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7bc11e209131cea8a32e.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa79ce1e209131cea7f7aa.md>)

Recorded evaluation configuration: {"Base Model Size":"4B","Training Dataset":"MATH-500"}. This record does not assert independent reproduction.

### TTRL

pass@1: 43.6. Result ID: 699ae4ac5698665ec474ca33.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7bc11e209131cea8a32e.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa79ce1e209131cea7f7aa.md>)

Recorded evaluation configuration: {"Base Model Size":"4B","Training Dataset":"AIME24"}. This record does not assert independent reproduction.

### TTRL

pass@1: 42.5. Result ID: 699ae4ac5698665ec474ca2f.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7bc11e209131cea8a32e.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa79ce1e209131cea7f7aa.md>)

Recorded evaluation configuration: {"Base Model Size":"4B","Training Dataset":"MATH-TRAIN"}. This record does not assert independent reproduction.

### Qwen3-4B-Base

pass@1: 39.3. Result ID: 699ae4ac5698665ec474ca2e.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7bc11e209131cea8a32e.md>)

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

Recorded evaluation configuration: {"Base Model Size":"4B","Training Dataset":"-"}. This record does not assert independent reproduction.

### Qwen3-8B-Base

pass@1: 38.7. Result ID: 699ae4ac5698665ec474ca35.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7bc11e209131cea8a32e.md>)

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

Recorded evaluation configuration: {"Base Model Size":"8B","Training Dataset":"-"}. This record does not assert independent reproduction.

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