Model comparison
Qwen3.6 Flash vs Qwen3.8 Max
Qwen3.8 Max is the stronger model overall, scoring 56.0 to 38.4 on the Noometry Index. Qwen3.6 Flash costs 7.1× less per token, which makes it the better buy when Qwen3.8 Max's lead doesn't matter for your workload.
Last verified . 9 shared benchmarks.
Alibaba (Qwen)Summary
- They share 9 benchmarks with published results for both. Qwen3.6 Flash scores higher in 0 categories and Qwen3.8 Max in 3 categories; 3 gaps are clear of the uncertainty.
- The widest gap is in math, where Qwen3.8 Max leads 73.1 to 38.0.
- The biggest single-benchmark swing is FrontierMath (Tiers 1-3): 22.5% for Qwen3.6 Flash and 74.7% for Qwen3.8 Max.
- Qwen3.6 Flash is cheaper at $0.19 / $1.13 per million input/output tokens, against $2 / $6 for Qwen3.8 Max.
Side by side
| Qwen3.6 Flash | Qwen3.8 Max | |
|---|---|---|
| Provider | Alibaba (Qwen) | Alibaba (Qwen) |
| Noometry Index | 38.4 | 56.0 |
| Released | 2026-04-27 | 2026-08-02 |
| Weights | Proprietary | Proprietary |
| Context window | 1M | 1M |
| Max output | 66K | 131K |
| Input $ / M tokens | $0.19 | $2 |
| Output $ / M tokens | $1.13 | $6 |
| Results tracked | 13 | 38 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Not comparable
Qwen3.6 Flash: —, Qwen3.8 Max: 52.3 (#30)
| Benchmark | Qwen3.6 Flash | Qwen3.8 Max |
|---|---|---|
| DeepSWE | — | 57.5% |
| LMArena WebDev | — | 1671 |
| FrontierSWE | — | 15.8% |
| SciCode | — | 52.9% |
| LMArena Coding | — | 1499 |
| ALE-Bench | 326.4 | — |
Agentic & Tool Use Not comparable
Qwen3.6 Flash: —, Qwen3.8 Max: 42.6 (#17)
| Benchmark | Qwen3.6 Flash | Qwen3.8 Max |
|---|---|---|
| τ²-bench Banking | — | 55.1% |
| GDP.pdf | — | 23.2% |
Reasoning Qwen3.8 Max leads
Qwen3.6 Flash: 28.4 (#90), Qwen3.8 Max: 53.4 (#26)
| Benchmark | Qwen3.6 Flash | Qwen3.8 Max |
|---|---|---|
| Chess Puzzles | 20% | 40% |
| Mystery Game Puzzles | 18% | 38% |
| DTBench | 77.1% | 92% |
| LMCA | 31% | 46.2% |
| Epoch Capabilities Index | 143.26 | 156.69 |
| SimpleBench | 35.2% | — |
| NYT Connections (extended) | — | 88.3% |
| CritPt | — | 20% |
| LMArena Hard Prompts | — | 1493 |
Math Qwen3.8 Max leads
Qwen3.6 Flash: 38.0 (#88), Qwen3.8 Max: 73.1 (#15)
| Benchmark | Qwen3.6 Flash | Qwen3.8 Max |
|---|---|---|
| FrontierMath (Tiers 1-3) | 22.5% | 74.7% |
| OTIS Mock AIME 2024-2025 | 84.4% | 100% |
| FrontierMath Tier 4 | — | 46.3% |
| ProofBench | — | 58% |
| LMArena Math | — | 1498 |
| FrontierMath (Feb 2025 set) | 10.3% | — |
| FrontierMath Tier 4 (v1) | 0% | — |
Knowledge Qwen3.8 Max leads
Qwen3.6 Flash: 42.2 (#90), Qwen3.8 Max: 61.7 (#25)
| Benchmark | Qwen3.6 Flash | Qwen3.8 Max |
|---|---|---|
| GPQA Diamond | 83.3% | 92.7% |
| SimpleQA Verified | 15.9% | 47.3% |
| LMArena Expert | — | 1504 |
Multimodal Not comparable
Qwen3.6 Flash: —, Qwen3.8 Max: 36.9 (#70)
| Benchmark | Qwen3.6 Flash | Qwen3.8 Max |
|---|---|---|
| LMArena Vision | — | 1314 |
| Furniture Assembly | — | 20% |
Multilingual Not comparable
Qwen3.6 Flash: —, Qwen3.8 Max: 56.6 (#18)
| Benchmark | Qwen3.6 Flash | Qwen3.8 Max |
|---|---|---|
| LMArena Non-English | — | 1468 |
| LMArena Chinese | — | 1537 |
| LMArena French | — | 1502 |
| LMArena German | — | 1489 |
| LMArena Japanese | — | 1466 |
| LMArena Korean | — | 1467 |
| LMArena Russian | — | 1478 |
| LMArena Spanish | — | 1495 |
Instruction Following Not comparable
Qwen3.6 Flash: —, Qwen3.8 Max: 77.7 (#19)
| Benchmark | Qwen3.6 Flash | Qwen3.8 Max |
|---|---|---|
| LMArena Instruction Following | — | 1478 |
Long Context Not comparable
Qwen3.6 Flash: —, Qwen3.8 Max: 45.5 (#30)
| Benchmark | Qwen3.6 Flash | Qwen3.8 Max |
|---|---|---|
| LMArena Longer Query | — | 1486 |
Writing & Preference Not comparable
Qwen3.6 Flash: —, Qwen3.8 Max: 67.0 (#31)
| Benchmark | Qwen3.6 Flash | Qwen3.8 Max |
|---|---|---|
| LMArena Text | — | 1480 |
| LMArena Creative Writing | — | 1478 |
| LMArena Multi-Turn | — | 1487 |
Frequently asked questions
Is Qwen3.6 Flash better than Qwen3.8 Max?
Qwen3.8 Max is the stronger model overall, scoring 56.0 to 38.4 on the Noometry Index. Qwen3.6 Flash costs 7.1× 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 Flash or Qwen3.8 Max?
Qwen3.6 Flash is cheaper. It lists at $0.19 per million input tokens and $1.13 per million output tokens; Qwen3.8 Max lists at $2 and $6.
Which has the bigger context window?
Both accept 1M tokens.
How many benchmarks do Qwen3.6 Flash and Qwen3.8 Max share?
9 benchmarks have published results for both models. Qwen3.6 Flash has 13 scored results on Noometry and Qwen3.8 Max has 38.