# Instruction Following Evaluation on Ours hard seed data

Source: [Instruction Following Evaluation on Ours hard seed data](<https://www.sota2.com/research/sota/instruction-following-evaluation-on-ours-hard-seed-data>)

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

Data updated: 2026-02-26T06:03:46.702Z

Evaluation of model performance on instruction following tasks using the proprietary 'Ours hard seed data' set.

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 | Instruction Following Evaluation |
| Dataset (including any stated split) | Ours hard seed data |
| Metric | Score |
| 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

- [Score](<https://www.sota2.com/research/sota/instruction-following-evaluation-on-ours-hard-seed-data.md?metric=Score>): Higher is better; unit: PERCENT.

- [Mean](<https://www.sota2.com/research/sota/instruction-following-evaluation-on-ours-hard-seed-data.md?metric=Mean>): Higher is better; unit: SCALAR.

- [Variance](<https://www.sota2.com/research/sota/instruction-following-evaluation-on-ours-hard-seed-data.md?metric=Variance>): Lower is better; unit: SCALAR.

## Recorded results

| Method | Score | Result date | Report paper |
| --- | --- | --- | --- |
| GPT-4 Turbo | 56.73 | Not recorded | 68fa75131e209131cea658ab |
| Claude3 | 53.75 | Not recorded | 68fa75131e209131cea658ab |
| GLM-4 | 51.75 | Not recorded | 68fa75131e209131cea658ab |
| Qwen | 49.85 | Not recorded | 68fa75131e209131cea658ab |
| GPT-4 | 47.51 | Not recorded | 68fa75131e209131cea658ab |

### GPT-4 Turbo

Score: 56.73. Result ID: 69938c413a6c31484916c3d9.

[Reporting paper](<https://www.sota2.com/research/paper/68fa75131e209131cea658ab.md>)

Recorded evaluation configuration: {"evaluation\_scale":"percentage-based"}. This record does not assert independent reproduction.

### Claude3

Score: 53.75. Result ID: 69938c413a6c31484916c3db.

[Reporting paper](<https://www.sota2.com/research/paper/68fa75131e209131cea658ab.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6c381e209131cea32de6.md>)

Recorded evaluation configuration: {"evaluation\_scale":"percentage-based"}. This record does not assert independent reproduction.

### GLM-4

Score: 51.75. Result ID: 69938c413a6c31484916c3d8.

[Reporting paper](<https://www.sota2.com/research/paper/68fa75131e209131cea658ab.md>)

Recorded evaluation configuration: {"evaluation\_scale":"percentage-based"}. This record does not assert independent reproduction.

### Qwen

Score: 49.85. Result ID: 69938c413a6c31484916c3dc.

[Reporting paper](<https://www.sota2.com/research/paper/68fa75131e209131cea658ab.md>)

Recorded evaluation configuration: {"evaluation\_scale":"percentage-based"}. This record does not assert independent reproduction.

### GPT-4

Score: 47.51. Result ID: 69938c413a6c31484916c3da.

[Reporting paper](<https://www.sota2.com/research/paper/68fa75131e209131cea658ab.md>)

Recorded evaluation configuration: {"evaluation\_scale":"percentage-based"}. This record does not assert independent reproduction.

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