Model comparison
Qwen3 235B-A22B vs Qwen3.6 Plus
Qwen3.6 Plus is the stronger model overall, scoring 47.5 to 43.5 on the Noometry Index.
Last verified . 30 shared benchmarks.
Alibaba (Qwen)Summary
- They share 30 benchmarks with published results for both. Qwen3 235B-A22B scores higher in 2 categories and Qwen3.6 Plus in 6 categories; 7 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where Qwen3.6 Plus leads 29.3 to 15.7.
- The biggest single-benchmark swing is GPQA Diamond: 80.1% for Qwen3 235B-A22B and 88.4% for Qwen3.6 Plus.
- Qwen3.6 Plus is cheaper at $0.50 / $3 per million input/output tokens, against $0.70 / $2.80 for Qwen3 235B-A22B.
- Qwen3.6 Plus accepts more context: 1M tokens versus 131K.
- Qwen3 235B-A22B has downloadable open weights; the other is API-only.
Side by side
| Qwen3 235B-A22B | Qwen3.6 Plus | |
|---|---|---|
| Provider | Alibaba (Qwen) | Alibaba (Qwen) |
| Noometry Index | 43.5 | 47.5 |
| Released | 2025-04 | 2026-03-31 |
| Weights | Open | Proprietary |
| Context window | 131K | 1M |
| Max output | 16K | 66K |
| Input $ / M tokens | $0.70 | $0.50 |
| Output $ / M tokens | $2.80 | $3 |
| Results tracked | 49 | 37 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Qwen3 235B-A22B leads
Qwen3 235B-A22B: 44.3 (#75), Qwen3.6 Plus: 40.8 (#130)
| Benchmark | Qwen3 235B-A22B | Qwen3.6 Plus |
|---|---|---|
| SciCode | 42.4% | 40.7% |
| LMArena Coding | 1445 | 1467 |
| SWE-bench Verified | — | 57.9% |
| Aider Polyglot | 59.6% | — |
| LMArena WebDev | — | 1461 |
| WeirdML | 41% | — |
| ALE-Bench | — | 670.15 |
Agentic & Tool Use Not comparable
Qwen3 235B-A22B: 33.9 (#51), Qwen3.6 Plus: —
| Benchmark | Qwen3 235B-A22B | Qwen3.6 Plus |
|---|---|---|
| Vending-Bench 2 | -11.34 | 5,115 |
| Berkeley Function Calling Leaderboard | 52.1% | — |
Reasoning Qwen3.6 Plus leads
Qwen3 235B-A22B: 15.7 (#311), Qwen3.6 Plus: 29.3 (#93)
| Benchmark | Qwen3 235B-A22B | Qwen3.6 Plus |
|---|---|---|
| CritPt | 0% | 2.9% |
| Chess Puzzles | 12% | 17% |
| LMArena Hard Prompts | 1433 | 1449 |
| Mystery Game Puzzles | 9% | 12% |
| DTBench | 80.3% | 81.9% |
| LMCA | 29.3% | 33.1% |
| Epoch Capabilities Index | 143.85 | 147.65 |
| ARC-AGI-2 | 1.3% | — |
| SimpleBench | 31% | — |
| Kagi LLM Benchmark | 69.4% | — |
| NYT Connections (extended) | — | 60.3% |
| ARC-AGI-1 | 11% | — |
| Thematic Generalization | — | 59.5% |
| ForecastBench | 59.7 | — |
Math Qwen3.6 Plus leads
Qwen3 235B-A22B: 50.4 (#57), Qwen3.6 Plus: 51.8 (#54)
| Benchmark | Qwen3 235B-A22B | Qwen3.6 Plus |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 86.7% | 93.3% |
| LMArena Math | 1432 | 1450 |
| FrontierMath (Feb 2025 set) | 8.5% | 26.2% |
| FrontierMath Tier 4 (v1) | 0% | 8.3% |
| FrontierMath (Tiers 1-3) | — | 38.2% |
| Omni-MATH | 71.8% | — |
| MATH Level 5 | 68.9% | — |
Knowledge Qwen3.6 Plus leads
Qwen3 235B-A22B: 49.6 (#73), Qwen3.6 Plus: 56.1 (#45)
| Benchmark | Qwen3 235B-A22B | Qwen3.6 Plus |
|---|---|---|
| GPQA Diamond | 80.1% | 88.4% |
| SimpleQA Verified | 40.4% | 44.1% |
| LMArena Expert | 1463 | 1454 |
| MMLU-Pro | 84.4% | — |
| Confabulations | 15.6% | — |
| Vectara Hallucination Rate | 9.3% | — |
| GPQA (HELM) | 72.7% | — |
Multilingual Qwen3.6 Plus leads
Qwen3 235B-A22B: 52.3 (#89), Qwen3.6 Plus: 53.3 (#70)
| Benchmark | Qwen3 235B-A22B | Qwen3.6 Plus |
|---|---|---|
| LMArena Non-English | 1409 | 1424 |
| LMArena Chinese | 1481 | 1477 |
| LMArena French | 1445 | 1455 |
| LMArena German | 1433 | 1452 |
| LMArena Japanese | 1399 | 1389 |
| LMArena Korean | 1391 | 1379 |
| LMArena Russian | 1411 | 1434 |
| LMArena Spanish | 1430 | 1432 |
Instruction Following Qwen3.6 Plus leads
Qwen3 235B-A22B: 72.6 (#136), Qwen3.6 Plus: 75.0 (#74)
| Benchmark | Qwen3 235B-A22B | Qwen3.6 Plus |
|---|---|---|
| LMArena Instruction Following | 1408 | 1425 |
| IFEval | 83.5% | — |
Long Context Too close to call
Qwen3 235B-A22B: 46.1 (#26), Qwen3.6 Plus: 45.2 (#49)
| Benchmark | Qwen3 235B-A22B | Qwen3.6 Plus |
|---|---|---|
| LMArena Longer Query | 1426 | 1439 |
| Fiction.LiveBench | 75% | — |
| CL-bench | — | 20.3% |
Writing & Preference Qwen3.6 Plus leads
Qwen3 235B-A22B: 59.6 (#108), Qwen3.6 Plus: 62.2 (#82)
| Benchmark | Qwen3 235B-A22B | Qwen3.6 Plus |
|---|---|---|
| LMArena Text | 1419 | 1437 |
| LMArena Creative Writing | 1384 | 1404 |
| LMArena Multi-Turn | 1432 | 1438 |
| Short-Story Creative Writing | 83% | — |
| EQ-Bench Creative Writing | 1366 | — |
| WildBench | 86.6% | — |
Frequently asked questions
Is Qwen3 235B-A22B better than Qwen3.6 Plus?
Qwen3.6 Plus is the stronger model overall, scoring 47.5 to 43.5 on the Noometry Index.
Which is cheaper, Qwen3 235B-A22B or Qwen3.6 Plus?
Qwen3.6 Plus is cheaper. It lists at $0.50 per million input tokens and $3 per million output tokens; Qwen3 235B-A22B lists at $0.70 and $2.80.
Is Qwen3 235B-A22B or Qwen3.6 Plus better for coding?
Qwen3 235B-A22B scores higher on coding benchmarks: 44.3 versus 40.8 in the Noometry coding category.
Which has the bigger context window?
Qwen3.6 Plus does, with 1M tokens against 131K.
How many benchmarks do Qwen3 235B-A22B and Qwen3.6 Plus share?
30 benchmarks have published results for both models. Qwen3 235B-A22B has 49 scored results on Noometry and Qwen3.6 Plus has 37.