# Mathematical Reasoning on AMC 2022, 2023, 2024

Source: [Mathematical Reasoning on AMC 2022, 2023, 2024](<https://www.sota2.com/research/sota/mathematical-reasoning-on-amc-2022-2023-2024>)

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

Data updated: 2026-05-08T07:49:10.580Z

Evaluation of model performance on mathematical reasoning tasks using the combined AMC 2022, 2023, and 2024 datasets, measured by Accuracy at 1 (Acc@1).

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 2022, 2023, 2024 |
| Metric | Accuracy@1 |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 15 of 15 |
| 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

- [Accuracy@1](<https://www.sota2.com/research/sota/mathematical-reasoning-on-amc-2022-2023-2024.md?metric=Accuracy%401>): Higher is better; unit: PERCENT.

## Recorded results

| Method | Accuracy@1 | Result date | Report paper |
| --- | --- | --- | --- |
| Qwen2.5-Math-7B + Resample w/ LOPE (w/ Training Signal Shaping) | 61.19 | Not recorded | 69fd6df75497b3b670c89346 |
| Qwen3-4B-Base + Resample w/ LOPE (w/ Training Signal Shaping) | 58.21 | Not recorded | 69fd6df75497b3b670c89346 |
| Qwen3-4B-Base + Resample w/ LOPE (w/o Training Signal Shaping) | 52.99 | Not recorded | 69fd6df75497b3b670c89346 |
| Qwen2.5-Math-7B + Resample w/ Naive Prompt | 50 | Not recorded | 69fd6df75497b3b670c89346 |
| Qwen3-4B-Base + GRPO | 47.76 | Not recorded | 69fd6df75497b3b670c89346 |
| Qwen2.5-Math-7B + GRPO | 47.76 | Not recorded | 69fd6df75497b3b670c89346 |
| Qwen2.5-Math-7B + Resample w/ LOPE (w/o Training Signal Shaping) | 47.01 | Not recorded | 69fd6df75497b3b670c89346 |
| Qwen3-4B-Base + Resample w/ Naive Prompt | 45.52 | Not recorded | 69fd6df75497b3b670c89346 |
| Qwen2.5-Math-7B | 35.4 | Not recorded | 69fd6df75497b3b670c89346 |
| Qwen3-1.7B-Base + Resample w/ LOPE (w/o Training Signal Shaping) | 33.58 | Not recorded | 69fd6df75497b3b670c89346 |
| Qwen3-1.7B-Base + Resample w/ LOPE (w/ Training Signal Shaping) | 32.84 | Not recorded | 69fd6df75497b3b670c89346 |
| Qwen3-4B-Base | 32.84 | Not recorded | 69fd6df75497b3b670c89346 |
| Qwen3-1.7B-Base + Resample w/ Naive Prompt | 28.36 | Not recorded | 69fd6df75497b3b670c89346 |
| Qwen3-1.7B-Base + GRPO | 27.61 | Not recorded | 69fd6df75497b3b670c89346 |
| Qwen3-1.7B-Base | 26.87 | Not recorded | 69fd6df75497b3b670c89346 |

### Qwen2.5-Math-7B + Resample w/ LOPE (w/ Training Signal Shaping)

Accuracy@1: 61.19. Result ID: 69fd957347f34ce924e05a44.

[Reporting paper](<https://www.sota2.com/research/paper/69fd6df75497b3b670c89346.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69fd6df75497b3b670c89346.md>)

Recorded evaluation configuration: {"Model":"Qwen2.5-Math-7B","Optimization Method":"GRPO + Resample","Prompt Strategy":"LOPE","Training Signal Shaping":"true"}. This record does not assert independent reproduction.

### Qwen3-4B-Base + Resample w/ LOPE (w/ Training Signal Shaping)

Accuracy@1: 58.21. Result ID: 69fd957347f34ce924e05a3f.

[Reporting paper](<https://www.sota2.com/research/paper/69fd6df75497b3b670c89346.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69fd6df75497b3b670c89346.md>)

Recorded evaluation configuration: {"Model":"Qwen3-4B-Base","Optimization Method":"GRPO + Resample","Prompt Strategy":"LOPE","Training Signal Shaping":"true"}. This record does not assert independent reproduction.

### Qwen3-4B-Base + Resample w/ LOPE (w/o Training Signal Shaping)

Accuracy@1: 52.99. Result ID: 69fd957347f34ce924e05a3e.

[Reporting paper](<https://www.sota2.com/research/paper/69fd6df75497b3b670c89346.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69fd6df75497b3b670c89346.md>)

Recorded evaluation configuration: {"Model":"Qwen3-4B-Base","Optimization Method":"GRPO + Resample","Prompt Strategy":"LOPE","Training Signal Shaping":"false"}. This record does not assert independent reproduction.

### Qwen2.5-Math-7B + Resample w/ Naive Prompt

Accuracy@1: 50. Result ID: 69fd957347f34ce924e05a42.

[Reporting paper](<https://www.sota2.com/research/paper/69fd6df75497b3b670c89346.md>)

Recorded evaluation configuration: {"Model":"Qwen2.5-Math-7B","Optimization Method":"GRPO + Resample","Prompt Strategy":"Naive Prompt"}. This record does not assert independent reproduction.

### Qwen3-4B-Base + GRPO

Accuracy@1: 47.76. Result ID: 69fd957347f34ce924e05a3c.

[Reporting paper](<https://www.sota2.com/research/paper/69fd6df75497b3b670c89346.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa73101e209131cea5a395.md>)

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

Recorded evaluation configuration: {"Model":"Qwen3-4B-Base","Optimization Method":"GRPO"}. This record does not assert independent reproduction.

### Qwen2.5-Math-7B + GRPO

Accuracy@1: 47.76. Result ID: 69fd957347f34ce924e05a41.

[Reporting paper](<https://www.sota2.com/research/paper/69fd6df75497b3b670c89346.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa73101e209131cea5a395.md>)

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

Recorded evaluation configuration: {"Model":"Qwen2.5-Math-7B","Optimization Method":"GRPO"}. This record does not assert independent reproduction.

### Qwen2.5-Math-7B + Resample w/ LOPE (w/o Training Signal Shaping)

Accuracy@1: 47.01. Result ID: 69fd957347f34ce924e05a43.

[Reporting paper](<https://www.sota2.com/research/paper/69fd6df75497b3b670c89346.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69fd6df75497b3b670c89346.md>)

Recorded evaluation configuration: {"Model":"Qwen2.5-Math-7B","Optimization Method":"GRPO + Resample","Prompt Strategy":"LOPE","Training Signal Shaping":"false"}. This record does not assert independent reproduction.

### Qwen3-4B-Base + Resample w/ Naive Prompt

Accuracy@1: 45.52. Result ID: 69fd957347f34ce924e05a3d.

[Reporting paper](<https://www.sota2.com/research/paper/69fd6df75497b3b670c89346.md>)

Recorded evaluation configuration: {"Model":"Qwen3-4B-Base","Optimization Method":"GRPO + Resample","Prompt Strategy":"Naive Prompt"}. This record does not assert independent reproduction.

### Qwen2.5-Math-7B

Accuracy@1: 35.4. Result ID: 69fd957347f34ce924e05a40.

[Reporting paper](<https://www.sota2.com/research/paper/69fd6df75497b3b670c89346.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa74d81e209131cea6419f.md>)

[Implementation](<https://github.com/QwenLM/Qwen2-Math>)

Recorded evaluation configuration: {"Model":"Qwen2.5-Math-7B","Optimization Method":"Base"}. This record does not assert independent reproduction.

### Qwen3-1.7B-Base + Resample w/ LOPE (w/o Training Signal Shaping)

Accuracy@1: 33.58. Result ID: 69fd957347f34ce924e05a39.

[Reporting paper](<https://www.sota2.com/research/paper/69fd6df75497b3b670c89346.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69fd6df75497b3b670c89346.md>)

Recorded evaluation configuration: {"Model":"Qwen3-1.7B-Base","Optimization Method":"GRPO + Resample","Prompt Strategy":"LOPE","Training Signal Shaping":"false"}. This record does not assert independent reproduction.

### Qwen3-1.7B-Base + Resample w/ LOPE (w/ Training Signal Shaping)

Accuracy@1: 32.84. Result ID: 69fd957347f34ce924e05a3a.

[Reporting paper](<https://www.sota2.com/research/paper/69fd6df75497b3b670c89346.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69fd6df75497b3b670c89346.md>)

Recorded evaluation configuration: {"Model":"Qwen3-1.7B-Base","Optimization Method":"GRPO + Resample","Prompt Strategy":"LOPE","Training Signal Shaping":"true"}. This record does not assert independent reproduction.

### Qwen3-4B-Base

Accuracy@1: 32.84. Result ID: 69fd957347f34ce924e05a3b.

[Reporting paper](<https://www.sota2.com/research/paper/69fd6df75497b3b670c89346.md>)

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

Recorded evaluation configuration: {"Model":"Qwen3-4B-Base","Optimization Method":"Base"}. This record does not assert independent reproduction.

### Qwen3-1.7B-Base + Resample w/ Naive Prompt

Accuracy@1: 28.36. Result ID: 69fd957347f34ce924e05a38.

[Reporting paper](<https://www.sota2.com/research/paper/69fd6df75497b3b670c89346.md>)

Recorded evaluation configuration: {"Model":"Qwen3-1.7B-Base","Optimization Method":"GRPO + Resample","Prompt Strategy":"Naive Prompt"}. This record does not assert independent reproduction.

### Qwen3-1.7B-Base + GRPO

Accuracy@1: 27.61. Result ID: 69fd957347f34ce924e05a37.

[Reporting paper](<https://www.sota2.com/research/paper/69fd6df75497b3b670c89346.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa73101e209131cea5a395.md>)

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

Recorded evaluation configuration: {"Model":"Qwen3-1.7B-Base","Optimization Method":"GRPO"}. This record does not assert independent reproduction.

### Qwen3-1.7B-Base

Accuracy@1: 26.87. Result ID: 69fd957347f34ce924e05a36.

[Reporting paper](<https://www.sota2.com/research/paper/69fd6df75497b3b670c89346.md>)

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

Recorded evaluation configuration: {"Model":"Qwen3-1.7B-Base","Optimization Method":"Base"}. This record does not assert independent reproduction.

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