Model comparison
Qwen3 32B vs Qwen3.5-Flash
Qwen3.5-Flash is the stronger model overall, scoring 42.3 to 38.8 on the Noometry Index.
Last verified . 20 shared benchmarks.
Alibaba (Qwen)Summary
- They share 20 benchmarks with published results for both. Qwen3 32B scores higher in 3 categories and Qwen3.5-Flash in 5 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where Qwen3.5-Flash leads 32.9 to 19.6.
- The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 66.9% for Qwen3 32B and 84.4% for Qwen3.5-Flash.
- Qwen3.5-Flash is cheaper at $0.10 / $0.40 per million input/output tokens, against $0.70 / $2.80 for Qwen3 32B.
- Qwen3.5-Flash accepts more context: 1M tokens versus 131K.
- Qwen3 32B has downloadable open weights; the other is API-only.
Side by side
| Qwen3 32B | Qwen3.5-Flash | |
|---|---|---|
| Provider | Alibaba (Qwen) | Alibaba (Qwen) |
| Noometry Index | 38.8 | 42.3 |
| Released | 2025-04 | 2026-02-23 |
| Weights | Open | Proprietary |
| Context window | 131K | 1M |
| Max output | 16K | 66K |
| Input $ / M tokens | $0.70 | $0.10 |
| Output $ / M tokens | $2.80 | $0.40 |
| Results tracked | 26 | 32 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Qwen3 32B leads
Qwen3 32B: 37.3 (#184), Qwen3.5-Flash: 33.9 (#234)
| Benchmark | Qwen3 32B | Qwen3.5-Flash |
|---|---|---|
| LMArena Coding | 1358 | 1412 |
| Aider Polyglot | 40% | — |
| LMArena WebDev | — | 1243 |
| SciCode | 35.4% | — |
| ALE-Bench | — | 221.8 |
Agentic & Tool Use Not comparable
Qwen3 32B: 32.3 (#58), Qwen3.5-Flash: —
| Benchmark | Qwen3 32B | Qwen3.5-Flash |
|---|---|---|
| Berkeley Function Calling Leaderboard | 48.7% | — |
| Vending-Bench 2 | — | 462.69 |
Reasoning Qwen3.5-Flash leads
Qwen3 32B: 19.6 (#233), Qwen3.5-Flash: 32.9 (#63)
| Benchmark | Qwen3 32B | Qwen3.5-Flash |
|---|---|---|
| Chess Puzzles | 5% | 21% |
| LMArena Hard Prompts | 1335 | 1404 |
| DTBench | 67.5% | 82.9% |
| LMCA | 17.3% | 29.1% |
| Epoch Capabilities Index | 138.51 | 143.98 |
| Kagi LLM Benchmark | 54.9% | — |
| CritPt | 0.3% | — |
| Mystery Game Puzzles | — | 20% |
Math Qwen3 32B leads
Qwen3 32B: 37.9 (#89), Qwen3.5-Flash: 36.3 (#117)
| Benchmark | Qwen3 32B | Qwen3.5-Flash |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 66.9% | 84.4% |
| LMArena Math | 1399 | 1407 |
| FrontierMath (Tiers 1-3) | — | 18.2% |
| FrontierMath (Feb 2025 set) | — | 6.2% |
| FrontierMath Tier 4 (v1) | — | 0% |
Knowledge Qwen3.5-Flash leads
Qwen3 32B: 40.0 (#112), Qwen3.5-Flash: 43.2 (#84)
| Benchmark | Qwen3 32B | Qwen3.5-Flash |
|---|---|---|
| GPQA Diamond | 65.7% | 82.3% |
| Vectara Hallucination Rate | 5.9% | 10.5% |
| LMArena Expert | 1362 | 1408 |
| SimpleQA Verified | — | 20.3% |
Multilingual Qwen3.5-Flash leads
Qwen3 32B: 45.8 (#162), Qwen3.5-Flash: 50.7 (#116)
| Benchmark | Qwen3 32B | Qwen3.5-Flash |
|---|---|---|
| LMArena Non-English | 1317 | 1385 |
| LMArena Chinese | 1357 | 1446 |
| LMArena German | 1341 | 1389 |
| LMArena Russian | 1311 | 1379 |
| LMArena French | — | 1412 |
| LMArena Japanese | — | 1369 |
| LMArena Korean | — | 1344 |
| LMArena Spanish | — | 1399 |
Instruction Following Qwen3.5-Flash leads
Qwen3 32B: 69.1 (#174), Qwen3.5-Flash: 72.8 (#133)
| Benchmark | Qwen3 32B | Qwen3.5-Flash |
|---|---|---|
| LMArena Instruction Following | 1305 | 1374 |
Long Context Qwen3 32B leads
Qwen3 32B: 43.7 (#83), Qwen3.5-Flash: 42.4 (#119)
| Benchmark | Qwen3 32B | Qwen3.5-Flash |
|---|---|---|
| LMArena Longer Query | 1327 | 1392 |
| Fiction.LiveBench | 74.2% | — |
Writing & Preference Qwen3.5-Flash leads
Qwen3 32B: 53.1 (#158), Qwen3.5-Flash: 58.2 (#117)
| Benchmark | Qwen3 32B | Qwen3.5-Flash |
|---|---|---|
| LMArena Text | 1340 | 1397 |
| LMArena Creative Writing | 1297 | 1343 |
| LMArena Multi-Turn | 1331 | 1393 |
Frequently asked questions
Is Qwen3 32B better than Qwen3.5-Flash?
Qwen3.5-Flash is the stronger model overall, scoring 42.3 to 38.8 on the Noometry Index.
Which is cheaper, Qwen3 32B 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; Qwen3 32B lists at $0.70 and $2.80.
Is Qwen3 32B or Qwen3.5-Flash better for coding?
Qwen3 32B scores higher on coding benchmarks: 37.3 versus 33.9 in the Noometry coding category.
Which has the bigger context window?
Qwen3.5-Flash does, with 1M tokens against 131K.
How many benchmarks do Qwen3 32B and Qwen3.5-Flash share?
20 benchmarks have published results for both models. Qwen3 32B has 26 scored results on Noometry and Qwen3.5-Flash has 32.