Model comparison
Qwen3.5-Flash vs Qwen3.8 Max
Qwen3.8 Max is the stronger model overall, scoring 56.8 to 42.5 on the Noometry Index. Qwen3.5-Flash costs 17× less per token, which makes it the better buy when Qwen3.8 Max's lead doesn't matter for your workload.
Last verified . 27 shared benchmarks.
Alibaba (Qwen)Summary
- They share 27 benchmarks with published results for both. Qwen3.5-Flash scores higher in 0 categories and Qwen3.8 Max in 8 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in math, where Qwen3.8 Max leads 73.2 to 37.4.
- The biggest single-benchmark swing is FrontierMath (Tiers 1-3): 18.2% for Qwen3.5-Flash and 74.7% for Qwen3.8 Max.
- Qwen3.5-Flash is cheaper at $0.10 / $0.40 per million input/output tokens, against $2 / $6 for Qwen3.8 Max.
Side by side
| Qwen3.5-Flash | Qwen3.8 Max | |
|---|---|---|
| Provider | Alibaba (Qwen) | Alibaba (Qwen) |
| Noometry Index | 42.5 | 56.8 |
| Released | 2026-02-23 | 2026-08-02 |
| Weights | Proprietary | Proprietary |
| Context window | 1M | 1M |
| Max output | 66K | 131K |
| Input $ / M tokens | $0.10 | $2 |
| Output $ / M tokens | $0.40 | $6 |
| Results tracked | 32 | 39 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Qwen3.8 Max leads
Qwen3.5-Flash: 34.2 (#242), Qwen3.8 Max: 53.5 (#29)
| Benchmark | Qwen3.5-Flash | Qwen3.8 Max |
|---|---|---|
| LMArena WebDev | 1244 | 1674 |
| LMArena Coding | 1412 | 1502 |
| DeepSWE | — | 57.5% |
| FrontierSWE | — | 17.8% |
| SciCode | — | 53.2% |
| ALE-Bench | 221.8 | — |
Agentic & Tool Use Not comparable
Qwen3.5-Flash: —, Qwen3.8 Max: 45.4 (#14)
| Benchmark | Qwen3.5-Flash | Qwen3.8 Max |
|---|---|---|
| APEX-Agents | — | 63.3% |
| τ²-bench Banking | — | 55.1% |
| GDP.pdf | — | 23.2% |
| Vending-Bench 2 | 462.69 | — |
Reasoning Qwen3.8 Max leads
Qwen3.5-Flash: 33.7 (#72), Qwen3.8 Max: 54.4 (#26)
| Benchmark | Qwen3.5-Flash | Qwen3.8 Max |
|---|---|---|
| Chess Puzzles | 21% | 40% |
| LMArena Hard Prompts | 1403 | 1496 |
| Mystery Game Puzzles | 20% | 38% |
| DTBench | 82.9% | 92% |
| LMCA | 29.1% | 46.2% |
| Epoch Capabilities Index | 143.98 | 156.41 |
| NYT Connections (extended) | — | 88.3% |
| CritPt | — | 20% |
Math Qwen3.8 Max leads
Qwen3.5-Flash: 37.4 (#158), Qwen3.8 Max: 73.2 (#20)
| Benchmark | Qwen3.5-Flash | Qwen3.8 Max |
|---|---|---|
| FrontierMath (Tiers 1-3) | 18.2% | 74.7% |
| OTIS Mock AIME 2024-2025 | 84.4% | 100% |
| LMArena Math | 1407 | 1499 |
| FrontierMath Tier 4 | — | 46.3% |
| ProofBench | — | 58% |
| FrontierMath (Feb 2025 set) | 6.2% | — |
| FrontierMath Tier 4 (v1) | 0% | — |
Knowledge Qwen3.8 Max leads
Qwen3.5-Flash: 43.2 (#93), Qwen3.8 Max: 61.7 (#27)
| Benchmark | Qwen3.5-Flash | Qwen3.8 Max |
|---|---|---|
| GPQA Diamond | 82.3% | 92.7% |
| SimpleQA Verified | 20.3% | 47.3% |
| LMArena Expert | 1407 | 1507 |
| Vectara Hallucination Rate | 10.5% | — |
Multimodal Not comparable
Qwen3.5-Flash: —, Qwen3.8 Max: 37.2 (#75)
| Benchmark | Qwen3.5-Flash | Qwen3.8 Max |
|---|---|---|
| LMArena Vision | — | 1314 |
| Furniture Assembly | — | 20% |
Multilingual Qwen3.8 Max leads
Qwen3.5-Flash: 50.5 (#121), Qwen3.8 Max: 56.7 (#18)
| Benchmark | Qwen3.5-Flash | Qwen3.8 Max |
|---|---|---|
| LMArena Non-English | 1385 | 1472 |
| LMArena Chinese | 1446 | 1538 |
| LMArena French | 1412 | 1503 |
| LMArena German | 1390 | 1483 |
| LMArena Japanese | 1368 | 1467 |
| LMArena Korean | 1344 | 1461 |
| LMArena Russian | 1379 | 1481 |
| LMArena Spanish | 1400 | 1492 |
Instruction Following Qwen3.8 Max leads
Qwen3.5-Flash: 72.6 (#139), Qwen3.8 Max: 77.6 (#17)
| Benchmark | Qwen3.5-Flash | Qwen3.8 Max |
|---|---|---|
| LMArena Instruction Following | 1374 | 1479 |
Long Context Qwen3.8 Max leads
Qwen3.5-Flash: 42.4 (#124), Qwen3.8 Max: 45.6 (#31)
| Benchmark | Qwen3.5-Flash | Qwen3.8 Max |
|---|---|---|
| LMArena Longer Query | 1392 | 1489 |
Writing & Preference Qwen3.8 Max leads
Qwen3.5-Flash: 57.9 (#122), Qwen3.8 Max: 67.1 (#30)
| Benchmark | Qwen3.5-Flash | Qwen3.8 Max |
|---|---|---|
| LMArena Text | 1397 | 1483 |
| LMArena Creative Writing | 1343 | 1479 |
| LMArena Multi-Turn | 1393 | 1489 |
Frequently asked questions
Is Qwen3.5-Flash better than Qwen3.8 Max?
Qwen3.8 Max is the stronger model overall, scoring 56.8 to 42.5 on the Noometry Index. Qwen3.5-Flash costs 17× less per token, which makes it the better buy when Qwen3.8 Max's lead doesn't matter for your workload.
Which is cheaper, Qwen3.5-Flash or Qwen3.8 Max?
Qwen3.5-Flash is cheaper. It lists at $0.10 per million input tokens and $0.40 per million output tokens; Qwen3.8 Max lists at $2 and $6.
Is Qwen3.5-Flash or Qwen3.8 Max better for coding?
Qwen3.8 Max scores higher on coding benchmarks: 53.5 versus 34.2 in the Noometry coding category.
Which has the bigger context window?
Both accept 1M tokens.
How many benchmarks do Qwen3.5-Flash and Qwen3.8 Max share?
27 benchmarks have published results for both models. Qwen3.5-Flash has 32 scored results on Noometry and Qwen3.8 Max has 39.