# Multi-turn consistency evaluation on MT-Consistency (test)

Source: [Multi-turn consistency evaluation on MT-Consistency (test)](<https://www.sota2.com/research/sota/multi-turn-consistency-evaluation-on-mt-consistency-test>)

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

Data updated: 2026-02-26T05:26:53.877Z

Evaluation of model performance on multi-turn conversational consistency using the MT-Consistency test set, measured by accuracy and specific consistency metrics (R\_sway, PWCScore).

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 | Multi-turn consistency evaluation |
| Dataset (including any stated split) | MT-Consistency (test) |
| Metric | Acc Avg |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 6 of 6 |
| 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

- [Acc Avg](<https://www.sota2.com/research/sota/multi-turn-consistency-evaluation-on-mt-consistency-test.md?metric=Acc%20Avg>): Higher is better; unit: PERCENT.

- [R\_sway](<https://www.sota2.com/research/sota/multi-turn-consistency-evaluation-on-mt-consistency-test.md?metric=R_sway>): Higher is better; unit: SCALAR.

- [PWCScore](<https://www.sota2.com/research/sota/multi-turn-consistency-evaluation-on-mt-consistency-test.md?metric=PWCScore>): Higher is better; unit: SCALAR.

## Recorded results

| Method | Acc Avg | Result date | Report paper |
| --- | --- | --- | --- |
| GPT | 71.34 | Not recorded | 68fa79d91e209131cea7fba0 |
| Claude | 63.07 | Not recorded | 68fa79d91e209131cea7fba0 |
| Qwen | 60.86 | Not recorded | 68fa79d91e209131cea7fba0 |
| Mistral | 50.02 | Not recorded | 68fa79d91e209131cea7fba0 |
| Gemini | 41.84 | Not recorded | 68fa79d91e209131cea7fba0 |
| LlaMa | 41.52 | Not recorded | 68fa79d91e209131cea7fba0 |

### GPT

Acc Avg: 71.34. Result ID: 698b628e7a967eddb76e28e6.

[Reporting paper](<https://www.sota2.com/research/paper/68fa79d91e209131cea7fba0.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa727a1e209131cea56a1e.md>)

[Implementation](<https://github.com/openai/evals>)

Recorded evaluation configuration: {"Setting":"Multi-Turn"}. This record does not assert independent reproduction.

### Claude

Acc Avg: 63.07. Result ID: 698b628e7a967eddb76e28e7.

[Reporting paper](<https://www.sota2.com/research/paper/68fa79d91e209131cea7fba0.md>)

Recorded evaluation configuration: {"Setting":"Multi-Turn"}. This record does not assert independent reproduction.

### Qwen

Acc Avg: 60.86. Result ID: 698b628e7a967eddb76e28e8.

[Reporting paper](<https://www.sota2.com/research/paper/68fa79d91e209131cea7fba0.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa767e1e209131cea6d6a7.md>)

Recorded evaluation configuration: {"Setting":"Multi-Turn"}. This record does not assert independent reproduction.

### Mistral

Acc Avg: 50.02. Result ID: 698b628e7a967eddb76e28eb.

[Reporting paper](<https://www.sota2.com/research/paper/68fa79d91e209131cea7fba0.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa70d61e209131cea4d909.md>)

[Implementation](<https://github.com/mistralai/mistral-src>)

Recorded evaluation configuration: {"Setting":"Multi-Turn"}. This record does not assert independent reproduction.

### Gemini

Acc Avg: 41.84. Result ID: 698b628e7a967eddb76e28e9.

[Reporting paper](<https://www.sota2.com/research/paper/68fa79d91e209131cea7fba0.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa76361e209131cea6be11.md>)

Recorded evaluation configuration: {"Setting":"Multi-Turn"}. This record does not assert independent reproduction.

### LlaMa

Acc Avg: 41.52. Result ID: 698b628e7a967eddb76e28ea.

[Reporting paper](<https://www.sota2.com/research/paper/68fa79d91e209131cea7fba0.md>)

Recorded evaluation configuration: {"Setting":"Multi-Turn"}. This record does not assert independent reproduction.

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