Model comparison
Claude Sonnet 4 vs Qwen3.5-Flash
Qwen3.5-Flash is the stronger model overall, scoring 42.5 to 40.8 on the Noometry Index.
Last verified . 26 shared benchmarks.
Alibaba (Qwen)Summary
- They share 26 benchmarks with published results for both. Claude Sonnet 4 scores higher in 2 categories and Qwen3.5-Flash in 6 categories; 6 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where Qwen3.5-Flash leads 33.7 to 22.9.
- The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 71.1% for Claude Sonnet 4 and 84.4% for Qwen3.5-Flash.
- Qwen3.5-Flash is cheaper at $0.10 / $0.40 per million input/output tokens, against $3 / $15 for Claude Sonnet 4.
- Qwen3.5-Flash accepts more context: 1M tokens versus 200K.
Side by side
| Claude Sonnet 4 | Qwen3.5-Flash | |
|---|---|---|
| Provider | Anthropic | Alibaba (Qwen) |
| Noometry Index | 40.8 | 42.5 |
| Released | 2025-05-22 | 2026-02-23 |
| Weights | Proprietary | Proprietary |
| Context window | 200K | 1M |
| Max output | 64K | 66K |
| Input $ / M tokens | $3 | $0.10 |
| Output $ / M tokens | $15 | $0.40 |
| Results tracked | 58 | 32 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Claude Sonnet 4 leads
Claude Sonnet 4: 43.5 (#88), Qwen3.5-Flash: 34.2 (#242)
| Benchmark | Claude Sonnet 4 | Qwen3.5-Flash |
|---|---|---|
| LMArena Coding | 1414 | 1412 |
| ALE-Bench | 655.35 | 221.8 |
| SWE-bench Verified (bash only) | 64.9% | — |
| Aider Polyglot | 61.3% | — |
| LMArena WebDev | — | 1244 |
| SciCode | 40% | — |
| GSO | 4.9% | — |
| WeirdML | 46.1% | — |
Agentic & Tool Use Not comparable
Claude Sonnet 4: 38.5 (#31), Qwen3.5-Flash: —
| Benchmark | Claude Sonnet 4 | Qwen3.5-Flash |
|---|---|---|
| TheAgentCompany | 33.1% | — |
| Cybench | 35% | — |
| DeepResearch Bench | 46.6% | — |
| OSWorld | 43.9% | — |
| METR Time Horizons | 62% | — |
| Vending-Bench 2 | — | 462.69 |
Reasoning Qwen3.5-Flash leads
Claude Sonnet 4: 22.9 (#187), Qwen3.5-Flash: 33.7 (#72)
| Benchmark | Claude Sonnet 4 | Qwen3.5-Flash |
|---|---|---|
| LMArena Hard Prompts | 1372 | 1403 |
| DTBench | 77.1% | 82.9% |
| LMCA | 29% | 29.1% |
| Epoch Capabilities Index | 141.69 | 143.98 |
| ARC-AGI-2 | 5.9% | — |
| SimpleBench | 45.5% | — |
| Kagi LLM Benchmark | 73% | — |
| ARC-AGI-1 | 40% | — |
| CritPt | 0.3% | — |
| Chess Puzzles | — | 21% |
| EnigmaEval | 3.1% | — |
| Mystery Game Puzzles | — | 20% |
| ForecastBench | 60.2 | — |
Math Claude Sonnet 4 leads
Claude Sonnet 4: 43.3 (#80), Qwen3.5-Flash: 37.4 (#158)
| Benchmark | Claude Sonnet 4 | Qwen3.5-Flash |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 71.1% | 84.4% |
| LMArena Math | 1375 | 1407 |
| FrontierMath (Feb 2025 set) | 4.1% | 6.2% |
| FrontierMath Tier 4 (v1) | 0% | 0% |
| FrontierMath (Tiers 1-3) | — | 18.2% |
| Omni-MATH | 60.2% | — |
| MATH Level 5 | 84.4% | — |
Knowledge Qwen3.5-Flash leads
Claude Sonnet 4: 41.8 (#108), Qwen3.5-Flash: 43.2 (#93)
| Benchmark | Claude Sonnet 4 | Qwen3.5-Flash |
|---|---|---|
| GPQA Diamond | 79.2% | 82.3% |
| Vectara Hallucination Rate | 10.3% | 10.5% |
| LMArena Expert | 1372 | 1407 |
| Humanity's Last Exam | 7.8% | — |
| SimpleQA Verified | — | 20.3% |
| MMLU-Pro | 84.3% | — |
| Confabulations | 13.2% | — |
| GPQA (HELM) | 70.6% | — |
Multimodal Not comparable
Claude Sonnet 4: 26.2 (#121), Qwen3.5-Flash: —
| Benchmark | Claude Sonnet 4 | Qwen3.5-Flash |
|---|---|---|
| LMArena Vision | 1191 | — |
| GeoBench | 37% | — |
| VPCT | 34% | — |
| MindCube | 44.8% | — |
Multilingual Qwen3.5-Flash leads
Claude Sonnet 4: 46.7 (#156), Qwen3.5-Flash: 50.5 (#121)
| Benchmark | Claude Sonnet 4 | Qwen3.5-Flash |
|---|---|---|
| LMArena Non-English | 1333 | 1385 |
| LMArena Chinese | 1350 | 1446 |
| LMArena French | 1363 | 1412 |
| LMArena German | 1331 | 1390 |
| LMArena Japanese | 1302 | 1368 |
| LMArena Korean | 1291 | 1344 |
| LMArena Russian | 1355 | 1379 |
| LMArena Spanish | 1357 | 1400 |
Instruction Following Too close to call
Claude Sonnet 4: 71.7 (#145), Qwen3.5-Flash: 72.6 (#139)
| Benchmark | Claude Sonnet 4 | Qwen3.5-Flash |
|---|---|---|
| LMArena Instruction Following | 1376 | 1374 |
| IFEval | 84% | — |
Long Context Qwen3.5-Flash leads
Claude Sonnet 4: 33.7 (#259), Qwen3.5-Flash: 42.4 (#124)
| Benchmark | Claude Sonnet 4 | Qwen3.5-Flash |
|---|---|---|
| LMArena Longer Query | 1398 | 1392 |
| Fiction.LiveBench | 46.9% | — |
Writing & Preference Too close to call
Claude Sonnet 4: 57.1 (#132), Qwen3.5-Flash: 57.9 (#122)
| Benchmark | Claude Sonnet 4 | Qwen3.5-Flash |
|---|---|---|
| LMArena Text | 1351 | 1397 |
| LMArena Creative Writing | 1345 | 1343 |
| LMArena Multi-Turn | 1376 | 1393 |
| Short-Story Creative Writing | 81.4% | — |
| EQ-Bench Creative Writing | 1483 | — |
| WildBench | 83.8% | — |
Frequently asked questions
Is Claude Sonnet 4 better than Qwen3.5-Flash?
Qwen3.5-Flash is the stronger model overall, scoring 42.5 to 40.8 on the Noometry Index.
Which is cheaper, Claude Sonnet 4 or Qwen3.5-Flash?
Qwen3.5-Flash is cheaper. It lists at $0.10 per million input tokens and $0.40 per million output tokens; Claude Sonnet 4 lists at $3 and $15.
Is Claude Sonnet 4 or Qwen3.5-Flash better for coding?
Claude Sonnet 4 scores higher on coding benchmarks: 43.5 versus 34.2 in the Noometry coding category.
Which has the bigger context window?
Qwen3.5-Flash does, with 1M tokens against 200K.
How many benchmarks do Claude Sonnet 4 and Qwen3.5-Flash share?
26 benchmarks have published results for both models. Claude Sonnet 4 has 58 scored results on Noometry and Qwen3.5-Flash has 32.