Benchmark
LVOmniBench
Scores are percentages. Higher is better.
Benchmark sourceScores from BenchBoard docs.
- Models measured
- 3
- Scores recorded
- 5
- Sources
- 1
- IQ weight
- No weight published
Score distribution
- 66.3%
- Mean
- 70.7%
- Median
- 7.4%
- Std dev
- 53.2%
- 10th percentile
- 73.6%
- 90th percentile
- 53.2%
- Min
Leaderboard
| # | Model | Score |
|---|---|---|
| 1 | Qwen3.8 Omni FlashAgent mode · Qwen Code· Reported by Qwen· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it. | 73.6% |
| 2 | Gemini 3.8 FlashAgent mode · Qwen Code· Reported by Qwen· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it. | 70.7% |
| 3 | Gemini 3.8 FlashStatic· Reported by Qwen· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it. | 70.7% |
| 4 | Qwen3.8 Omni FlashStatic· Reported by Qwen· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it. | 63.3% |
| 5 | Qwen3.5 Omni PlusStatic· Reported by Qwen· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it. | 53.2% |