# Skill Routing on Skill Routing Dataset Hard

Source: [Skill Routing on Skill Routing Dataset Hard](<https://www.sota2.com/research/sota/skill-routing-on-skill-routing-dataset-hard>)

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

Data updated: 2026-03-25T14:26:50.180Z

Evaluation of Skill Routing performance on the Skill Routing Dataset Hard, measured by Hit@1 and MRR.

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 | Skill Routing |
| Dataset (including any stated split) | Skill Routing Dataset Hard |
| Metric | Hit@1 |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 26 of 26 |
| 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

- [Hit@1](<https://www.sota2.com/research/sota/skill-routing-on-skill-routing-dataset-hard.md?metric=Hit%401>): Higher is better; unit: PERCENT.

- [MRR](<https://www.sota2.com/research/sota/skill-routing-on-skill-routing-dataset-hard.md?metric=MRR>): Higher is better; unit: PERCENT.

## Recorded results

| Method | Hit@1 | Result date | Report paper |
| --- | --- | --- | --- |
| SR-Emb-8B + SR-Rank-8B | 73.3 | Not recorded | 69c36bef0636e5bf4afc637c |
| SR-Emb-0.6B + SR-Rank-0.6B | 72 | Not recorded | 69c36bef0636e5bf4afc637c |
| SR-Emb-0.6B + Qwen3-Rank-8B | 70.7 | Not recorded | 69c36bef0636e5bf4afc637c |
| SR-Emb-0.6B + SR-Rank-8B | 70.7 | Not recorded | 69c36bef0636e5bf4afc637c |
| SR-Emb-0.6B + Qwen3-Rank-0.6B | 69.3 | Not recorded | 69c36bef0636e5bf4afc637c |
| Qwen3-Emb-8B + Qwen3-Rank-8B | 68 | Not recorded | 69c36bef0636e5bf4afc637c |
| Qwen3-Emb-0.6B + Qwen3-Rank-8B | 68 | Not recorded | 69c36bef0636e5bf4afc637c |
| SR-Emb-0.6B + GPT-5.4-mini | 65.3 | Not recorded | 69c36bef0636e5bf4afc637c |
| SR-Emb-0.6B + GPT-4o-mini | 65.3 | Not recorded | 69c36bef0636e5bf4afc637c |
| SR-Emb-0.6B | 64 | Not recorded | 69c36bef0636e5bf4afc637c |
| Qwen3-Emb-8B | 62.7 | Not recorded | 69c36bef0636e5bf4afc637c |
| Qwen3-Emb-8B + GPT-4o-mini | 62.7 | Not recorded | 69c36bef0636e5bf4afc637c |
| Qwen3-Emb-0.6B + Qwen3-Rank-0.6B | 62.7 | Not recorded | 69c36bef0636e5bf4afc637c |
| Qwen3-Emb-8B + GPT-5.4-mini | 56 | Not recorded | 69c36bef0636e5bf4afc637c |
| Qwen3-Emb-0.6B + GPT-4o-mini | 54.7 | Not recorded | 69c36bef0636e5bf4afc637c |
| Qwen3-Emb-0.6B + GPT-5.4-mini | 54.7 | Not recorded | 69c36bef0636e5bf4afc637c |
| Qwen3-Emb-8B + Qwen3-Rank-0.6B | 54.7 | Not recorded | 69c36bef0636e5bf4afc637c |
| Qwen3-Emb-0.6B | 53.3 | Not recorded | 69c36bef0636e5bf4afc637c |
| Qwen3-Emb-0.6B + GPT-5.4-mini | 29.3 | Not recorded | 69c36bef0636e5bf4afc637c |
| Qwen3-Emb-8B + GPT-5.4-mini | 26.7 | Not recorded | 69c36bef0636e5bf4afc637c |
| Qwen3-Emb-0.6B + Qwen3-Rank-8B | 25.3 | Not recorded | 69c36bef0636e5bf4afc637c |
| Qwen3-Emb-8B + Qwen3-Rank-8B | 18.7 | Not recorded | 69c36bef0636e5bf4afc637c |
| Qwen3-Emb-0.6B + Qwen3-Rank-0.6B | 17.3 | Not recorded | 69c36bef0636e5bf4afc637c |
| Qwen3-Emb-8B + GPT-4o-mini | 17.3 | Not recorded | 69c36bef0636e5bf4afc637c |
| Qwen3-Emb-0.6B + GPT-4o-mini | 16 | Not recorded | 69c36bef0636e5bf4afc637c |
| Qwen3-Emb-8B + Qwen3-Rank-0.6B | 13.3 | Not recorded | 69c36bef0636e5bf4afc637c |

### SR-Emb-8B + SR-Rank-8B

Hit@1: 73.3. Result ID: 69c3ef4047f34ce924adbdb8.

[Reporting paper](<https://www.sota2.com/research/paper/69c36bef0636e5bf4afc637c.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69c36bef0636e5bf4afc637c.md>)

Recorded evaluation configuration: {"Encoder":"SR-Emb-8B","Reranker":"SR-Rank-8B","Rank Input":"full"}. This record does not assert independent reproduction.

### SR-Emb-0.6B + SR-Rank-0.6B

Hit@1: 72. Result ID: 69c3ef4047f34ce924adbdb6.

[Reporting paper](<https://www.sota2.com/research/paper/69c36bef0636e5bf4afc637c.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69c36bef0636e5bf4afc637c.md>)

Recorded evaluation configuration: {"Encoder":"SR-Emb-0.6B","Reranker":"SR-Rank-0.6B","Rank Input":"full"}. This record does not assert independent reproduction.

### SR-Emb-0.6B + Qwen3-Rank-8B

Hit@1: 70.7. Result ID: 69c3ef4047f34ce924adbdb5.

[Reporting paper](<https://www.sota2.com/research/paper/69c36bef0636e5bf4afc637c.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69c36bef0636e5bf4afc637c.md>)

Recorded evaluation configuration: {"Encoder":"SR-Emb-0.6B","Reranker":"Qwen3-Rank-8B","Rank Input":"full"}. This record does not assert independent reproduction.

### SR-Emb-0.6B + SR-Rank-8B

Hit@1: 70.7. Result ID: 69c3ef4047f34ce924adbdb7.

[Reporting paper](<https://www.sota2.com/research/paper/69c36bef0636e5bf4afc637c.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69c36bef0636e5bf4afc637c.md>)

Recorded evaluation configuration: {"Encoder":"SR-Emb-0.6B","Reranker":"SR-Rank-8B","Rank Input":"full"}. This record does not assert independent reproduction.

### SR-Emb-0.6B + Qwen3-Rank-0.6B

Hit@1: 69.3. Result ID: 69c3ef4047f34ce924adbdb4.

[Reporting paper](<https://www.sota2.com/research/paper/69c36bef0636e5bf4afc637c.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69c36bef0636e5bf4afc637c.md>)

Recorded evaluation configuration: {"Encoder":"SR-Emb-0.6B","Reranker":"Qwen3-Rank-0.6B","Rank Input":"full"}. This record does not assert independent reproduction.

### Qwen3-Emb-8B + Qwen3-Rank-8B

Hit@1: 68. Result ID: 69c3ef4047f34ce924adbdb0.

[Reporting paper](<https://www.sota2.com/research/paper/69c36bef0636e5bf4afc637c.md>)

Recorded evaluation configuration: {"Encoder":"Qwen3-Emb-8B","Reranker":"Qwen3-Rank-8B","Rank Input":"full"}. This record does not assert independent reproduction.

### Qwen3-Emb-0.6B + Qwen3-Rank-8B

Hit@1: 68. Result ID: 69c3ef4047f34ce924adbdb1.

[Reporting paper](<https://www.sota2.com/research/paper/69c36bef0636e5bf4afc637c.md>)

Recorded evaluation configuration: {"Encoder":"Qwen3-Emb-0.6B","Reranker":"Qwen3-Rank-8B","Rank Input":"full"}. This record does not assert independent reproduction.

### SR-Emb-0.6B + GPT-5.4-mini

Hit@1: 65.3. Result ID: 69c3ef4047f34ce924adbdb2.

[Reporting paper](<https://www.sota2.com/research/paper/69c36bef0636e5bf4afc637c.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69c36bef0636e5bf4afc637c.md>)

Recorded evaluation configuration: {"Encoder":"SR-Emb-0.6B","Reranker":"GPT-5.4-mini","Rank Input":"full"}. This record does not assert independent reproduction.

### SR-Emb-0.6B + GPT-4o-mini

Hit@1: 65.3. Result ID: 69c3ef4047f34ce924adbdb3.

[Reporting paper](<https://www.sota2.com/research/paper/69c36bef0636e5bf4afc637c.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69c36bef0636e5bf4afc637c.md>)

Recorded evaluation configuration: {"Encoder":"SR-Emb-0.6B","Reranker":"GPT-4o-mini","Rank Input":"full"}. This record does not assert independent reproduction.

### SR-Emb-0.6B

Hit@1: 64. Result ID: 69c3ef4047f34ce924adbda1.

[Reporting paper](<https://www.sota2.com/research/paper/69c36bef0636e5bf4afc637c.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69c36bef0636e5bf4afc637c.md>)

Recorded evaluation configuration: {"Encoder":"SR-Emb-0.6B","Reranker":"None","Rank Input":"N/A"}. This record does not assert independent reproduction.

### Qwen3-Emb-8B

Hit@1: 62.7. Result ID: 69c3ef4047f34ce924adbda0.

[Reporting paper](<https://www.sota2.com/research/paper/69c36bef0636e5bf4afc637c.md>)

Recorded evaluation configuration: {"Encoder":"Qwen3-Emb-8B","Reranker":"None","Rank Input":"N/A"}. This record does not assert independent reproduction.

### Qwen3-Emb-8B + GPT-4o-mini

Hit@1: 62.7. Result ID: 69c3ef4047f34ce924adbdaa.

[Reporting paper](<https://www.sota2.com/research/paper/69c36bef0636e5bf4afc637c.md>)

Recorded evaluation configuration: {"Encoder":"Qwen3-Emb-8B","Reranker":"GPT-4o-mini","Rank Input":"full"}. This record does not assert independent reproduction.

### Qwen3-Emb-0.6B + Qwen3-Rank-0.6B

Hit@1: 62.7. Result ID: 69c3ef4047f34ce924adbdae.

[Reporting paper](<https://www.sota2.com/research/paper/69c36bef0636e5bf4afc637c.md>)

Recorded evaluation configuration: {"Encoder":"Qwen3-Emb-0.6B","Reranker":"Qwen3-Rank-0.6B","Rank Input":"full"}. This record does not assert independent reproduction.

### Qwen3-Emb-8B + GPT-5.4-mini

Hit@1: 56. Result ID: 69c3ef4047f34ce924adbdac.

[Reporting paper](<https://www.sota2.com/research/paper/69c36bef0636e5bf4afc637c.md>)

Recorded evaluation configuration: {"Encoder":"Qwen3-Emb-8B","Reranker":"GPT-5.4-mini","Rank Input":"full"}. This record does not assert independent reproduction.

### Qwen3-Emb-0.6B + GPT-4o-mini

Hit@1: 54.7. Result ID: 69c3ef4047f34ce924adbdab.

[Reporting paper](<https://www.sota2.com/research/paper/69c36bef0636e5bf4afc637c.md>)

Recorded evaluation configuration: {"Encoder":"Qwen3-Emb-0.6B","Reranker":"GPT-4o-mini","Rank Input":"full"}. This record does not assert independent reproduction.

### Qwen3-Emb-0.6B + GPT-5.4-mini

Hit@1: 54.7. Result ID: 69c3ef4047f34ce924adbdad.

[Reporting paper](<https://www.sota2.com/research/paper/69c36bef0636e5bf4afc637c.md>)

Recorded evaluation configuration: {"Encoder":"Qwen3-Emb-0.6B","Reranker":"GPT-5.4-mini","Rank Input":"full"}. This record does not assert independent reproduction.

### Qwen3-Emb-8B + Qwen3-Rank-0.6B

Hit@1: 54.7. Result ID: 69c3ef4047f34ce924adbdaf.

[Reporting paper](<https://www.sota2.com/research/paper/69c36bef0636e5bf4afc637c.md>)

Recorded evaluation configuration: {"Encoder":"Qwen3-Emb-8B","Reranker":"Qwen3-Rank-0.6B","Rank Input":"full"}. This record does not assert independent reproduction.

### Qwen3-Emb-0.6B

Hit@1: 53.3. Result ID: 69c3ef4047f34ce924adbd9f.

[Reporting paper](<https://www.sota2.com/research/paper/69c36bef0636e5bf4afc637c.md>)

Recorded evaluation configuration: {"Encoder":"Qwen3-Emb-0.6B","Reranker":"None","Rank Input":"N/A"}. This record does not assert independent reproduction.

### Qwen3-Emb-0.6B + GPT-5.4-mini

Hit@1: 29.3. Result ID: 69c3ef4047f34ce924adbda9.

[Reporting paper](<https://www.sota2.com/research/paper/69c36bef0636e5bf4afc637c.md>)

Recorded evaluation configuration: {"Encoder":"Qwen3-Emb-0.6B","Reranker":"GPT-5.4-mini","Rank Input":"nd"}. This record does not assert independent reproduction.

### Qwen3-Emb-8B + GPT-5.4-mini

Hit@1: 26.7. Result ID: 69c3ef4047f34ce924adbda8.

[Reporting paper](<https://www.sota2.com/research/paper/69c36bef0636e5bf4afc637c.md>)

Recorded evaluation configuration: {"Encoder":"Qwen3-Emb-8B","Reranker":"GPT-5.4-mini","Rank Input":"nd"}. This record does not assert independent reproduction.

### Qwen3-Emb-0.6B + Qwen3-Rank-8B

Hit@1: 25.3. Result ID: 69c3ef4047f34ce924adbda4.

[Reporting paper](<https://www.sota2.com/research/paper/69c36bef0636e5bf4afc637c.md>)

Recorded evaluation configuration: {"Encoder":"Qwen3-Emb-0.6B","Reranker":"Qwen3-Rank-8B","Rank Input":"nd"}. This record does not assert independent reproduction.

### Qwen3-Emb-8B + Qwen3-Rank-8B

Hit@1: 18.7. Result ID: 69c3ef4047f34ce924adbda2.

[Reporting paper](<https://www.sota2.com/research/paper/69c36bef0636e5bf4afc637c.md>)

Recorded evaluation configuration: {"Encoder":"Qwen3-Emb-8B","Reranker":"Qwen3-Rank-8B","Rank Input":"nd"}. This record does not assert independent reproduction.

### Qwen3-Emb-0.6B + Qwen3-Rank-0.6B

Hit@1: 17.3. Result ID: 69c3ef4047f34ce924adbda3.

[Reporting paper](<https://www.sota2.com/research/paper/69c36bef0636e5bf4afc637c.md>)

Recorded evaluation configuration: {"Encoder":"Qwen3-Emb-0.6B","Reranker":"Qwen3-Rank-0.6B","Rank Input":"nd"}. This record does not assert independent reproduction.

### Qwen3-Emb-8B + GPT-4o-mini

Hit@1: 17.3. Result ID: 69c3ef4047f34ce924adbda6.

[Reporting paper](<https://www.sota2.com/research/paper/69c36bef0636e5bf4afc637c.md>)

Recorded evaluation configuration: {"Encoder":"Qwen3-Emb-8B","Reranker":"GPT-4o-mini","Rank Input":"nd"}. This record does not assert independent reproduction.

### Qwen3-Emb-0.6B + GPT-4o-mini

Hit@1: 16. Result ID: 69c3ef4047f34ce924adbda7.

[Reporting paper](<https://www.sota2.com/research/paper/69c36bef0636e5bf4afc637c.md>)

Recorded evaluation configuration: {"Encoder":"Qwen3-Emb-0.6B","Reranker":"GPT-4o-mini","Rank Input":"nd"}. This record does not assert independent reproduction.

### Qwen3-Emb-8B + Qwen3-Rank-0.6B

Hit@1: 13.3. Result ID: 69c3ef4047f34ce924adbda5.

[Reporting paper](<https://www.sota2.com/research/paper/69c36bef0636e5bf4afc637c.md>)

Recorded evaluation configuration: {"Encoder":"Qwen3-Emb-8B","Reranker":"Qwen3-Rank-0.6B","Rank Input":"nd"}. This record does not assert independent reproduction.

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