# Formal Theorem Proving on HS Competition (10)

Source: [Formal Theorem Proving on HS Competition (10)](<https://www.sota2.com/research/sota/formal-theorem-proving-on-hs-competition-10>)

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

Data updated: 2026-04-30T07:13:49.023Z

Evaluation of model performance on formal theorem proving tasks using the HS Competition (10) dataset, measured by Proof Length.

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 | Formal Theorem Proving |
| Dataset (including any stated split) | HS Competition (10) |
| Metric | Proof Length |
| Unit | SCALAR |
| Direction | Lower is better |
| Results loaded | 5 of 5 |
| 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

- [Proof Length](<https://www.sota2.com/research/sota/formal-theorem-proving-on-hs-competition-10.md?metric=Proof%20Length>): Lower is better; unit: SCALAR.

## Recorded results

| Method | Proof Length | Result date | Report paper |
| --- | --- | --- | --- |
| DreamProver (Gemini 3.1 Pro) | 17.8 | Not recorded | 69f2e22955a3c32af89fc71f |
| DreamProver (GPT-5.3-Codex) | 22.3 | Not recorded | 69f2e22955a3c32af89fc71f |
| DreamProver (Gemini 2.5 Pro) | 23.3 | Not recorded | 69f2e22955a3c32af89fc71f |
| GPT-5.3-Codex | 32 | Not recorded | 69f2e22955a3c32af89fc71f |
| Gemini 3.1 Pro | 41.5 | Not recorded | 69f2e22955a3c32af89fc71f |

### DreamProver (Gemini 3.1 Pro)

Proof Length: 17.8. Result ID: 69f3012847f34ce924d8fdcb.

[Reporting paper](<https://www.sota2.com/research/paper/69f2e22955a3c32af89fc71f.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69f2e22955a3c32af89fc71f.md>)

Recorded evaluation configuration: {"Method Category":"Lemma Learning System","Backbone Model":"Gemini 3.1 Pro"}. This record does not assert independent reproduction.

### DreamProver (GPT-5.3-Codex)

Proof Length: 22.3. Result ID: 69f3012847f34ce924d8fdc9.

[Reporting paper](<https://www.sota2.com/research/paper/69f2e22955a3c32af89fc71f.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69f2e22955a3c32af89fc71f.md>)

Recorded evaluation configuration: {"Method Category":"Lemma Learning System","Backbone Model":"GPT-5.3-Codex"}. This record does not assert independent reproduction.

### DreamProver (Gemini 2.5 Pro)

Proof Length: 23.3. Result ID: 69f3012847f34ce924d8fdca.

[Reporting paper](<https://www.sota2.com/research/paper/69f2e22955a3c32af89fc71f.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69f2e22955a3c32af89fc71f.md>)

Recorded evaluation configuration: {"Method Category":"Lemma Learning System","Backbone Model":"Gemini 2.5 Pro"}. This record does not assert independent reproduction.

### GPT-5.3-Codex

Proof Length: 32. Result ID: 69f3012847f34ce924d8fdc7.

[Reporting paper](<https://www.sota2.com/research/paper/69f2e22955a3c32af89fc71f.md>)

Recorded evaluation configuration: {"Method Category":"Proprietary LLMs"}. This record does not assert independent reproduction.

### Gemini 3.1 Pro

Proof Length: 41.5. Result ID: 69f3012847f34ce924d8fdc8.

[Reporting paper](<https://www.sota2.com/research/paper/69f2e22955a3c32af89fc71f.md>)

Recorded evaluation configuration: {"Method Category":"Proprietary LLMs"}. This record does not assert independent reproduction.

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