- MMLU-Redux 86.6
- GPQA 69.6
- AIME25 65
- ZebraLogic 76
- MultiPL-E 81.4
- IFEval 81
- Creative Writing v3 80.6
- WritingBench 82.6
- BFCL-v3 64.4
- MultiIF 64
- PolyMATH 37.9
- LiveBench 20241125 (Thinking variant) 71.8
- Arena-Hard v2 (Thinking variant) 55.1
- Mean of Librispeech test-clean 1.22 / test-other 2.48 1.85
- Mean of CV15-en 6.05 / CV15-zh 4.31 5.18
- Fleurs-avg (19 lang) ASR WER; Fleurs-en 2.72, Fleurs-zh 2.20 5.33
- VoiceBench Overall 85.5
- MMAU-v05.15.25 77.5
- RUL-MuchoMusic 52
- MMStar 68.5
- HallusionBench 59.7
- MM-MT-Bench 7.4
- MMMU_val 69.1
- MMMU_pro 57
- MathVista_mini 75.9
- MathVision_full 56.3
- AI2D 85.2
- ChartQA_test 86.8
- CountBench 90
- Video-MME 70.5
- LVBench 50.2
- MLVU 75.2
- WorldSense 54
- DailyOmni (Thinking variant) 75.8
- VideoHolmes (Thinking variant) 57.3
- SEED test-zh / test-en 1.07 / 1.39 (test-en recorded) 1.39
- Mean of Wenetspeech net 4.69 / meeting 5.89 5.29
Qwen3-Omni-30B-A3B
Qwen3-Omni-30B-A3B.
38
Benchmarks with a score
0
Ranked first, out of five or more models
0
In the top three
Qwen Sep 2025 Qwen3-Omni-30B-A3B
Open weights Mid-size
No Epoch Capabilities Index score yet
- MMMU-Pro 57% #60 /72
- OCRBench v2 61.3% #8 /65
- HallusionBench 59.7% #20 /23
Where it ranks
Its best score on each benchmark, and where that ranks among every model with a score there. We rank each model once, at its best setting.
Knowledge and language
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| WritingBench | 82.6 | #4 of 14 | Model card | |
| Creative Writing v3 | 80.6 | #6 of 12 | Model card | |
| Multi-IF | 64% | #20 of 28 | Model card | |
| MMLU-Redux | 86.6% | #28 of 38 | Model card | |
| Arena-Hard v2.0 | 55.1% | #18 of 24 | Model card | |
| IFEval | 81% | #96 of 126 | Model card | |
| HallusionBench | 59.7% | #20 of 23 | Model card |
Math and science
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| GPQA Diamond | 69.6% | #174 of 284 | Model card | |
| ZebraLogic | 76% | #19 of 23 | Model card | |
| AIME 2025 | 65% | #119 of 143 | Model card | |
| PolyMath | 37.9% | #17 of 18 | Model card |
Coding
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| MultiPL-E | 81.4% | #4 of 18 | Model card |
Agents and tools
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| Berkeley Function Calling Leaderboard (BFCL) | 64.4% | #35 of 47 | Model card |
Vision, audio and robotics
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| OCRBench v2 | 61.3% | #8 of 65 | Official leaderboard | |
| MMAU | 77.5% | #4 of 14 | Model card | |
| MathVista | 75.9% | #29 of 64 | Model card | |
| ChartQA | 86.8% | #18 of 37 | Model card | |
| VideoMME v2 (with subtitles) | 19.5% | #19 of 31 | Official leaderboard | |
| Audio MC | 24.3% | #10 of 16 | Official leaderboard | |
| CountBench | 90% | #13 of 20 | Model card | |
| MMMU | 69.1% | #64 of 97 | Model card | |
| MM-MT-Bench | 7.40 | #4 of 6 | Model card | |
| VideoDR (with search) | 37% | #4 of 6 | Official leaderboard | |
| FLEURS | 5.33% | #2 of 3 | Model card | |
| VoiceBench | 85.5% | #7 of 10 | Model card | |
| AI2D | 85.2% | #32 of 44 | Model card | |
| MMSU | 69% | #4 of 5 | Model card | |
| MATH-Vision (MathVision) | 56.3% | #34 of 42 | Model card | |
| MLVU | 75.2% | #13 of 16 | Model card | |
| MMMU-Pro | 57% | #60 of 72 | Model card | |
| MMStar | 68.5% | #25 of 30 | Model card | |
| Daily-Omni | 75.8% | #5 of 6 | Model card | |
| Video-MME | 70.5% | #37 of 42 | Model card | |
| LVBench | 50.2% | #23 of 25 | Model card | |
| Video-Holmes | 57.3% | #6 of 6 | Model card | |
| WorldSense (omnimodal) | 54% | #5 of 5 | Model card | |
| MuChoMusic | 52% | #3 of 3 | Model card |
Other
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| LiveBench | 71.8% | #10 of 27 | Model card |
Its launch post
The benchmarks the lab chose to headline, with the scores it gave.
Its system card
Every result in the longer documents the lab published with it. 3 of them aren't in the launch post.
Qwen3-Omni Technical Report 22 Sep 2025
Capabilities
- MMLU-Redux 86.6
- GPQA Diamond 69.6
- AIME 2025 65
- ZebraLogic 76
- MultiPL-E 81.4
- IFEval 81
- Creative Writing v3 80.6
- WritingBench 82.6
- Berkeley Function Calling Leaderboard (BFCL) 64.4
- Multi-IF 64
- PolyMath 37.9
- LiveBench 71.8
- Arena-Hard v2.0 55.1
- WenetSpeech
- LibriSpeech
- Common Voice
- FLEURS 5.33
- VoiceBench 85.5
- MMAU 77.5
- MMSU 69
- MuChoMusic 52
- MMStar 68.5
- HallusionBench 59.7
- MM-MT-Bench 7.4
- MMMU 69.1
- MMMU-Pro 57
- MathVista 75.9
- MATH-Vision (MathVision) 56.3
- AI2D 85.2
- ChartQA 86.8
- CountBench 90
- Video-MME 70.5
- LVBench 50.2
- MLVU 75.2
- WorldSense (omnimodal) 54
- Daily-Omni 75.8
- Video-Holmes 57.3
- Seed-TTS-eval
- MiniMax TTS Multilingual Test Set
- CV3-Eval