Qwen 3.5 Plus (hosted 397B-A17B)
Qwen 3.5 Plus (hosted 397B-A17B).
26
Benchmarks with a score
0
Ranked first, out of five or more models
0
In the top three
Qwen Feb 2026 Qwen 3.5 Plus (hosted 397B-A17B)
147.3
#82 of 256
Epoch Capabilities Index
- Toolathlon 37.7% #25 /50
- GPQA Diamond 84.8% #81 /229
- Vals Finance Agent 54.5% #17 /49
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 | |
|---|---|---|---|---|
| CL-bench | 19.8% | #9 of 24 | Official leaderboard | |
| CL-bench Life | 12.4% | #10 of 16 | Official leaderboard | |
| SimpleQA Verified | 25.4% | #76 of 94 | Epoch AI run |
Math and science
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| GPQA Diamond | 84.8% | #81 of 229 | Epoch AI run | |
| OTIS Mock AIME 2024-2025 | 86.7% | #80 of 212 | Epoch AI run | |
| FrontierMath | 21% | #29 of 70 | Epoch AI run | |
| DTBench | 80.5% | #89 of 180 | Official leaderboard | |
| FrontierMath Tier 4 | 2.08% | #44 of 55 | Epoch AI run |
Coding
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| WebDev Arena | 1,399 | #66 of 105 | arena.ai, via Epoch AI | |
| Vibe Code Bench | 15.7% | #94 of 120 | Official leaderboard |
Agents and tools
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| Toolathlon | 37.7% | #25 of 50 | Official leaderboard | |
| UniClawBench | 31.8% | #6 of 10 | Official leaderboard | |
| Mystery Game Puzzles | 17% | #59 of 84 | Epoch AI run | |
| Chess Puzzles | 9.00% | #87 of 93 | Epoch AI run |
Professional work
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| LegalBench | 85.1% | #26 of 138 | Official leaderboard | |
| CorpFin v2 | 65.3% | #30 of 119 | Official leaderboard | |
| Vals Finance Agent | 54.5% | #17 of 49 | Official leaderboard | |
| CaseLaw v2 | 59.7% | #26 of 53 | Official leaderboard | |
| Vals Index v1 | 50.8% | #23 of 34 | Official leaderboard | |
| MortgageTax | 60.8% | #61 of 87 | Official leaderboard | |
| Vending-Bench 2 | $1 | #63 of 67 | Official leaderboard | |
| E-Commerce Bench | 0.01 | #19 of 19 | Official leaderboard |
Other
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| LMCA | 36.4% | #75 of 147 | Official leaderboard | |
| Conceptual Reasoning Index | 46.5 | #81 of 151 | Official leaderboard | |
| ACCoRD | 56.1 | #102 of 152 | Official leaderboard | |
| SAGE (Vals) | 30.4% | #75 of 81 | Official leaderboard |
Qwen's models over time
Qwen's models in color, against every other model on the Epoch Capabilities Index, by release date.
Source: Epoch Capabilities Index, Epoch AI (CC BY 4.0)