Model comparison
Qwen3.6 Plus vs Qwen3.8 Max
Qwen3.8 Max is the stronger model overall, scoring 56.8 to 47.5 on the Noometry Index. Qwen3.6 Plus costs 2.7× less per token, which makes it the better buy when Qwen3.8 Max's lead doesn't matter for your workload.
Last verified . 30 shared benchmarks.
Alibaba (Qwen)Summary
- They share 30 benchmarks with published results for both. Qwen3.6 Plus scores higher in 0 categories and Qwen3.8 Max in 8 categories; 7 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where Qwen3.8 Max leads 54.4 to 29.3.
- The biggest single-benchmark swing is FrontierMath (Tiers 1-3): 38.2% for Qwen3.6 Plus and 74.7% for Qwen3.8 Max.
- Qwen3.6 Plus is cheaper at $0.50 / $3 per million input/output tokens, against $2 / $6 for Qwen3.8 Max.
Side by side
| Qwen3.6 Plus | Qwen3.8 Max | |
|---|---|---|
| Provider | Alibaba (Qwen) | Alibaba (Qwen) |
| Noometry Index | 47.5 | 56.8 |
| Released | 2026-03-31 | 2026-08-02 |
| Weights | Proprietary | Proprietary |
| Context window | 1M | 1M |
| Max output | 66K | 131K |
| Input $ / M tokens | $0.50 | $2 |
| Output $ / M tokens | $3 | $6 |
| Results tracked | 37 | 39 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Qwen3.8 Max leads
Qwen3.6 Plus: 40.8 (#130), Qwen3.8 Max: 53.5 (#29)
| Benchmark | Qwen3.6 Plus | Qwen3.8 Max |
|---|---|---|
| LMArena WebDev | 1461 | 1674 |
| SciCode | 40.7% | 53.2% |
| LMArena Coding | 1467 | 1502 |
| SWE-bench Verified | 57.9% | — |
| DeepSWE | — | 57.5% |
| FrontierSWE | — | 17.8% |
| ALE-Bench | 670.15 | — |
Agentic & Tool Use Not comparable
Qwen3.6 Plus: —, Qwen3.8 Max: 45.4 (#14)
| Benchmark | Qwen3.6 Plus | Qwen3.8 Max |
|---|---|---|
| APEX-Agents | — | 63.3% |
| τ²-bench Banking | — | 55.1% |
| GDP.pdf | — | 23.2% |
| Vending-Bench 2 | 5,115 | — |
Reasoning Qwen3.8 Max leads
Qwen3.6 Plus: 29.3 (#93), Qwen3.8 Max: 54.4 (#26)
| Benchmark | Qwen3.6 Plus | Qwen3.8 Max |
|---|---|---|
| NYT Connections (extended) | 60.3% | 88.3% |
| CritPt | 2.9% | 20% |
| Chess Puzzles | 17% | 40% |
| LMArena Hard Prompts | 1449 | 1496 |
| Mystery Game Puzzles | 12% | 38% |
| DTBench | 81.9% | 92% |
| LMCA | 33.1% | 46.2% |
| Epoch Capabilities Index | 147.65 | 156.41 |
| Thematic Generalization | 59.5% | — |
Math Qwen3.8 Max leads
Qwen3.6 Plus: 51.8 (#54), Qwen3.8 Max: 73.2 (#20)
| Benchmark | Qwen3.6 Plus | Qwen3.8 Max |
|---|---|---|
| FrontierMath (Tiers 1-3) | 38.2% | 74.7% |
| OTIS Mock AIME 2024-2025 | 93.3% | 100% |
| LMArena Math | 1450 | 1499 |
| FrontierMath Tier 4 | — | 46.3% |
| ProofBench | — | 58% |
| FrontierMath (Feb 2025 set) | 26.2% | — |
| FrontierMath Tier 4 (v1) | 8.3% | — |
Knowledge Qwen3.8 Max leads
Qwen3.6 Plus: 56.1 (#45), Qwen3.8 Max: 61.7 (#27)
| Benchmark | Qwen3.6 Plus | Qwen3.8 Max |
|---|---|---|
| GPQA Diamond | 88.4% | 92.7% |
| SimpleQA Verified | 44.1% | 47.3% |
| LMArena Expert | 1454 | 1507 |
Multimodal Not comparable
Qwen3.6 Plus: —, Qwen3.8 Max: 37.2 (#75)
| Benchmark | Qwen3.6 Plus | Qwen3.8 Max |
|---|---|---|
| LMArena Vision | — | 1314 |
| Furniture Assembly | — | 20% |
Multilingual Qwen3.8 Max leads
Qwen3.6 Plus: 53.3 (#70), Qwen3.8 Max: 56.7 (#18)
| Benchmark | Qwen3.6 Plus | Qwen3.8 Max |
|---|---|---|
| LMArena Non-English | 1424 | 1472 |
| LMArena Chinese | 1477 | 1538 |
| LMArena French | 1455 | 1503 |
| LMArena German | 1452 | 1483 |
| LMArena Japanese | 1389 | 1467 |
| LMArena Korean | 1379 | 1461 |
| LMArena Russian | 1434 | 1481 |
| LMArena Spanish | 1432 | 1492 |
Instruction Following Qwen3.8 Max leads
Qwen3.6 Plus: 75.0 (#74), Qwen3.8 Max: 77.6 (#17)
| Benchmark | Qwen3.6 Plus | Qwen3.8 Max |
|---|---|---|
| LMArena Instruction Following | 1425 | 1479 |
Long Context Too close to call
Qwen3.6 Plus: 45.2 (#49), Qwen3.8 Max: 45.6 (#31)
| Benchmark | Qwen3.6 Plus | Qwen3.8 Max |
|---|---|---|
| LMArena Longer Query | 1439 | 1489 |
| CL-bench | 20.3% | — |
Writing & Preference Qwen3.8 Max leads
Qwen3.6 Plus: 62.2 (#82), Qwen3.8 Max: 67.1 (#30)
| Benchmark | Qwen3.6 Plus | Qwen3.8 Max |
|---|---|---|
| LMArena Text | 1437 | 1483 |
| LMArena Creative Writing | 1404 | 1479 |
| LMArena Multi-Turn | 1438 | 1489 |
Frequently asked questions
Is Qwen3.6 Plus better than Qwen3.8 Max?
Qwen3.8 Max is the stronger model overall, scoring 56.8 to 47.5 on the Noometry Index. Qwen3.6 Plus costs 2.7× less per token, which makes it the better buy when Qwen3.8 Max's lead doesn't matter for your workload.
Which is cheaper, Qwen3.6 Plus or Qwen3.8 Max?
Qwen3.6 Plus is cheaper. It lists at $0.50 per million input tokens and $3 per million output tokens; Qwen3.8 Max lists at $2 and $6.
Is Qwen3.6 Plus or Qwen3.8 Max better for coding?
Qwen3.8 Max scores higher on coding benchmarks: 53.5 versus 40.8 in the Noometry coding category.
Which has the bigger context window?
Both accept 1M tokens.
How many benchmarks do Qwen3.6 Plus and Qwen3.8 Max share?
30 benchmarks have published results for both models. Qwen3.6 Plus has 37 scored results on Noometry and Qwen3.8 Max has 39.