Model comparison
Gemini 2.5 Flash vs Qwen3 235B-A22B
Qwen3 235B-A22B is the stronger model overall, scoring 43.2 to 39.0 on the Noometry Index.
Last verified . 43 shared benchmarks.
Google
Alibaba (Qwen)Summary
- They share 43 benchmarks with published results for both. Gemini 2.5 Flash scores higher in 3 categories and Qwen3 235B-A22B in 6 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in knowledge, where Qwen3 235B-A22B leads 49.5 to 36.1.
- The biggest single-benchmark swing is GPQA (HELM): 39% for Gemini 2.5 Flash and 72.7% for Qwen3 235B-A22B.
- Gemini 2.5 Flash is cheaper at $0.30 / $2.50 per million input/output tokens, against $0.70 / $2.80 for Qwen3 235B-A22B.
- Gemini 2.5 Flash accepts more context: 1.05M tokens versus 131K.
- Qwen3 235B-A22B has downloadable open weights; the other is API-only.
Side by side
| Gemini 2.5 Flash | Qwen3 235B-A22B | |
|---|---|---|
| Provider | Alibaba (Qwen) | |
| Noometry Index | 39.0 | 43.2 |
| Released | 2025-04-17 | 2025-04 |
| Weights | Proprietary | Open |
| Context window | 1.05M | 131K |
| Max output | 66K | 16K |
| Input $ / M tokens | $0.30 | $0.70 |
| Output $ / M tokens | $2.50 | $2.80 |
| Results tracked | 54 | 49 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Qwen3 235B-A22B leads
Gemini 2.5 Flash: 35.5 (#210), Qwen3 235B-A22B: 43.8 (#72)
| Benchmark | Gemini 2.5 Flash | Qwen3 235B-A22B |
|---|---|---|
| Aider Polyglot | 55.1% | 59.6% |
| WeirdML | 41.9% | 41% |
| LMArena Coding | 1424 | 1445 |
| SWE-bench Verified (bash only) | 28.7% | — |
| SciCode | — | 42.4% |
| ALE-Bench | 661.88 | — |
Agentic & Tool Use Qwen3 235B-A22B leads
Gemini 2.5 Flash: 30.6 (#67), Qwen3 235B-A22B: 33.7 (#48)
| Benchmark | Gemini 2.5 Flash | Qwen3 235B-A22B |
|---|---|---|
| Berkeley Function Calling Leaderboard | 56.2% | 52.1% |
| Vending-Bench 2 | 548.84 | -11.34 |
| Terminal-Bench | 17.1% | — |
| TheAgentCompany | 41.1% | — |
| BALROG | 33.5% | — |
Reasoning Gemini 2.5 Flash leads
Gemini 2.5 Flash: 18.2 (#271), Qwen3 235B-A22B: 15.8 (#302)
| Benchmark | Gemini 2.5 Flash | Qwen3 235B-A22B |
|---|---|---|
| ARC-AGI-2 | 2.5% | 1.3% |
| SimpleBench | 41.2% | 31% |
| Kagi LLM Benchmark | 56.8% | 69.4% |
| ARC-AGI-1 | 33.3% | 11% |
| CritPt | 1.1% | 0% |
| LMArena Hard Prompts | 1422 | 1433 |
| DTBench | 76.5% | 80.3% |
| LMCA | 27.5% | 29.3% |
| Epoch Capabilities Index | 143.04 | 143.87 |
| ForecastBench | 60.6 | 59.7 |
| Chess Puzzles | — | 12% |
| EnigmaEval | 2.7% | — |
| Mystery Game Puzzles | — | 9% |
Math Qwen3 235B-A22B leads
Gemini 2.5 Flash: 38.1 (#87), Qwen3 235B-A22B: 48.5 (#51)
| Benchmark | Gemini 2.5 Flash | Qwen3 235B-A22B |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 73.1% | 86.7% |
| Omni-MATH | 38.5% | 71.8% |
| LMArena Math | 1415 | 1431 |
| FrontierMath (Feb 2025 set) | 4.8% | 8.5% |
| FrontierMath Tier 4 (v1) | 4.2% | 0% |
| MATH Level 5 | — | 68.9% |
Knowledge Qwen3 235B-A22B leads
Gemini 2.5 Flash: 36.1 (#162), Qwen3 235B-A22B: 49.5 (#67)
| Benchmark | Gemini 2.5 Flash | Qwen3 235B-A22B |
|---|---|---|
| MMLU-Pro | 63.9% | 84.4% |
| Confabulations | 16.8% | 15.6% |
| Vectara Hallucination Rate | 7.8% | 9.3% |
| GPQA (HELM) | 39% | 72.7% |
| LMArena Expert | 1426 | 1460 |
| GPQA Diamond | — | 80.1% |
| Humanity's Last Exam | 12.1% | — |
| SimpleQA Verified | — | 40.4% |
Multimodal Not comparable
Gemini 2.5 Flash: 41.2 (#31), Qwen3 235B-A22B: —
| Benchmark | Gemini 2.5 Flash | Qwen3 235B-A22B |
|---|---|---|
| LMArena Vision | 1254 | — |
| GeoBench | 76% | — |
| VPCT | 46.2% | — |
| SpatialViz-Bench | 36.9% | — |
Multilingual Too close to call
Gemini 2.5 Flash: 52.5 (#83), Qwen3 235B-A22B: 52.5 (#84)
| Benchmark | Gemini 2.5 Flash | Qwen3 235B-A22B |
|---|---|---|
| LMArena Non-English | 1410 | 1410 |
| LMArena Chinese | 1450 | 1481 |
| LMArena French | 1433 | 1445 |
| LMArena German | 1419 | 1432 |
| LMArena Japanese | 1403 | 1400 |
| LMArena Korean | 1385 | 1391 |
| LMArena Russian | 1415 | 1411 |
| LMArena Spanish | 1421 | 1429 |
Instruction Following Gemini 2.5 Flash leads
Gemini 2.5 Flash: 75.9 (#50), Qwen3 235B-A22B: 72.8 (#132)
| Benchmark | Gemini 2.5 Flash | Qwen3 235B-A22B |
|---|---|---|
| IFEval | 89.8% | 83.5% |
| LMArena Instruction Following | 1405 | 1408 |
Long Context Gemini 2.5 Flash leads
Gemini 2.5 Flash: 47.5 (#16), Qwen3 235B-A22B: 46.1 (#25)
| Benchmark | Gemini 2.5 Flash | Qwen3 235B-A22B |
|---|---|---|
| Fiction.LiveBench | 77.8% | 75% |
| LMArena Longer Query | 1419 | 1426 |
Writing & Preference Qwen3 235B-A22B leads
Gemini 2.5 Flash: 54.0 (#153), Qwen3 235B-A22B: 59.8 (#103)
| Benchmark | Gemini 2.5 Flash | Qwen3 235B-A22B |
|---|---|---|
| LMArena Text | 1417 | 1419 |
| LMArena Creative Writing | 1400 | 1384 |
| Short-Story Creative Writing | 76.5% | 83% |
| EQ-Bench Creative Writing | 1137 | 1366 |
| WildBench | 81.7% | 86.6% |
| LMArena Multi-Turn | 1408 | 1432 |
Frequently asked questions
Is Gemini 2.5 Flash better than Qwen3 235B-A22B?
Qwen3 235B-A22B is the stronger model overall, scoring 43.2 to 39.0 on the Noometry Index.
Which is cheaper, Gemini 2.5 Flash or Qwen3 235B-A22B?
Gemini 2.5 Flash is cheaper. It lists at $0.30 per million input tokens and $2.50 per million output tokens; Qwen3 235B-A22B lists at $0.70 and $2.80.
Is Gemini 2.5 Flash or Qwen3 235B-A22B better for coding?
Qwen3 235B-A22B scores higher on coding benchmarks: 43.8 versus 35.5 in the Noometry coding category.
Which has the bigger context window?
Gemini 2.5 Flash does, with 1.05M tokens against 131K.
How many benchmarks do Gemini 2.5 Flash and Qwen3 235B-A22B share?
43 benchmarks have published results for both models. Gemini 2.5 Flash has 54 scored results on Noometry and Qwen3 235B-A22B has 49.