- MMLU-Pro 82.7
- MMLU-Redux 92.5
- GPQA 77.2
- SuperGPQA 60.8
- AIME25 87.8
- HMMT25 73.9
- LiveBench 241125 76.6
- LiveCodeBench v6 (25.02-25.05) 68.7
- CFEval 2071
- OJBench 29.7
- IFEval 88.9
- Arena-Hard v2* (win rate judged by GPT-4.1) 62.3
- WritingBench 84.6
- BFCL-v3 72
- Mean of TAU1-Retail 69.6 / TAU1-Airline 49.0 59.3
- Mean of TAU2-Retail 67.8 / TAU2-Airline 60.5 / TAU2-Telecom 43.9 57.4
- MultiIF 77.8
- MMLU-ProX 78.7
- INCLUDE 78.9
- PolyMATH 56.3
Qwen3-Next-80B-A3B
Qwen3-Next-80B-A3B.
46
Benchmarks with a score
0
Ranked first, out of five or more models
0
In the top three
Qwen Sep 2025 Qwen3-Next-80B-A3B
Open weights Mid-size
137.3
#203 of 404
Self-Reported Index, from the lab's own numbers. No Epoch Capabilities Index score yet.
- SuperGPQA 60.8% #19 /35
- AMO-Bench 34.8% #8 /21
- KINA 28.3% #25 /37
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 | |
|---|---|---|---|---|
| Multi-IF | 77.8% | #8 of 28 | Lab model card | |
| MMLU-Pro | 82.7% | #53 of 167 | Lab model card | |
| IFEval | 88.9% | #47 of 135 | Lab model card | |
| WritingBench | 84.6 | #5 of 14 | Lab model card | |
| Vectara Hallucination Leaderboard (HHEM) | 3.70% | #60 of 148 | Official leaderboard | |
| MMLU-Redux | 92.5% | #20 of 45 | Lab model card | |
| INCLUDE | 78.9% | #14 of 31 | Lab model card | |
| LMArena (Elo) | 1,369 | #161 of 335 | Official leaderboard | |
| Vectara Hallucination Leaderboard (2025 dataset) | 9.30% | #50 of 104 | Official leaderboard | |
| SuperGPQA | 60.8% | #19 of 35 | Lab model card | |
| MMLU-ProX | 78.7% | #13 of 23 | Lab model card | |
| Arena-Hard v2.0 | 62.3% | #16 of 24 | Lab model card | |
| KINA | 28.3% | #25 of 37 | Official leaderboard |
Math and science
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| GPQA Main | 63% | #21 of 60 | Official leaderboard | |
| AMO-Bench | 34.8% | #8 of 21 | Official leaderboard | |
| AIME 2025 | 87.8% | #62 of 143 | Lab model card | |
| GPQA Diamond | 77.2% | #135 of 302 | Lab model card | |
| ChemBench | 61% | #11 of 24 | Official leaderboard | |
| PolyMath | 56.3% | #13 of 24 | Lab model card | |
| DTBench | 75.2% | #106 of 180 | Official leaderboard | |
| HMMT 2025 | 73.9% | #43 of 54 | Lab model card |
Coding
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| Codeforces rating | 2,071 | #22 of 47 | Lab model card | |
| LiveCodeBench v6 | 68.7% | #59 of 90 | Lab model card | |
| OJBench | 29.7% | #11 of 16 | Lab model card | |
| BigCodeArena | 902 | #12 of 13 | Official leaderboard |
Agents and tools
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| Berkeley Function Calling Leaderboard (BFCL) | 72% | #9 of 41 | Lab model card | |
| SnakeBench | 26 | #63 of 188 | Official leaderboard | |
| tau-bench | 59.3% | #12 of 23 | Lab model card | |
| LLM Chess | -128 | #81 of 123 | Official leaderboard | |
| Pencil Puzzle Bench | 0% | #38 of 53 | Official leaderboard | |
| tau2-bench | 57.4% | #68 of 79 | Lab model card |
Professional work
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| BRIDGE | 50.9 | #5 of 39 | Official leaderboard | |
| Physical-Engineering's Last Exam | 0.04 | #17 of 18 | Official leaderboard |
Safety and security
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| Anthropic Red Team | 99.9% | #4 of 75 | Official leaderboard | |
| AIR-Bench 2024 | 86.7% | #8 of 82 | Official leaderboard | |
| HELM Safety | 94.8 | #24 of 75 | Official leaderboard | |
| SimpleSafetyTests | 99.5% | #27 of 75 | Official leaderboard | |
| Human Pathogen Capabilities Test | 47.2% | #55 of 86 | Official leaderboard | |
| nocot-bench (NCRI) | 63.8 | #125 of 185 | Official leaderboard | |
| Molecular Biology Capabilities Test | 42.2% | #70 of 90 | Official leaderboard |
Other
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| LiveBench | 76.6% | #6 of 27 | Lab model card | |
| HELM Capabilities | 70.0 | #21 of 60 | Official leaderboard | |
| ACCoRD | 59.6 | #91 of 152 | Official leaderboard | |
| Political Science LLM Benchmark | 65.1% | #29 of 42 | Official leaderboard | |
| Conceptual Reasoning Index | 36.7 | #105 of 151 | Official leaderboard | |
| LMCA | 21.8% | #109 of 147 | Official leaderboard |
Its lab's numbers put it at 137.3, number 203 of 404
The Self-Reported Index fits one score to the 17 capability results Qwen reported for this model, on the scale of the Epoch Capabilities Index. Labs choose which benchmarks to show and how to run them, so read it as the lab's claim, not an independent measurement. The 90% interval runs from 135.6 to 139.1. How we compute it.
- MMLU-Redux 92.5
- IFEval 88.9
- AIME 2025 87.8
- MMLU-Pro 82.7
- INCLUDE 78.9
- MMLU-ProX 78.7
- Multi-IF 77.8
- GPQA Diamond 77.2
- HMMT 2025 73.9
- Berkeley Function Calling Leaderboard (BFCL) 72
- LiveCodeBench v6 68.7
- Arena-Hard v2.0 62.3
- SuperGPQA 60.8
- tau-bench 59.3
- tau2-bench 57.4
- PolyMath 56.3
- OJBench 29.7
Its lab's number next to a leaderboard
Where the benchmark's own leaderboard, or Epoch AI's run, also has this model. Gaps mostly come from different harnesses, effort levels and task subsets. Every pair.
Its launch post
The benchmarks the lab chose to headline, with the scores it gave.