Model comparison
Gemini 2.5 Flash vs Qwen3.8 Max
Qwen3.8 Max is the stronger model overall, scoring 56.8 to 39.3 on the Noometry Index. Gemini 2.5 Flash costs 3.5× less per token, which makes it the better buy when Qwen3.8 Max's lead doesn't matter for your workload.
Last verified . 23 shared benchmarks.
Google
Alibaba (Qwen)Summary
- They share 23 benchmarks with published results for both. Gemini 2.5 Flash scores higher in 2 categories and Qwen3.8 Max in 8 categories; 10 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where Qwen3.8 Max leads 54.4 to 18.1.
- The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 73.1% for Gemini 2.5 Flash and 100% for Qwen3.8 Max.
- Gemini 2.5 Flash is cheaper at $0.30 / $2.50 per million input/output tokens, against $2 / $6 for Qwen3.8 Max.
- Gemini 2.5 Flash accepts more context: 1.05M tokens versus 1M.
Side by side
| Gemini 2.5 Flash | Qwen3.8 Max | |
|---|---|---|
| Provider | Alibaba (Qwen) | |
| Noometry Index | 39.3 | 56.8 |
| Released | 2025-04-17 | 2026-08-02 |
| Weights | Proprietary | Proprietary |
| Context window | 1.05M | 1M |
| Max output | 66K | 131K |
| Input $ / M tokens | $0.30 | $2 |
| Output $ / M tokens | $2.50 | $6 |
| Results tracked | 54 | 39 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Qwen3.8 Max leads
Gemini 2.5 Flash: 35.8 (#220), Qwen3.8 Max: 53.5 (#29)
| Benchmark | Gemini 2.5 Flash | Qwen3.8 Max |
|---|---|---|
| LMArena Coding | 1424 | 1502 |
| DeepSWE | — | 57.5% |
| SWE-bench Verified (bash only) | 28.7% | — |
| Aider Polyglot | 55.1% | — |
| LMArena WebDev | — | 1674 |
| FrontierSWE | — | 17.8% |
| SciCode | — | 53.2% |
| WeirdML | 41.9% | — |
| ALE-Bench | 661.88 | — |
Agentic & Tool Use Qwen3.8 Max leads
Gemini 2.5 Flash: 30.8 (#74), Qwen3.8 Max: 45.4 (#14)
| Benchmark | Gemini 2.5 Flash | Qwen3.8 Max |
|---|---|---|
| Terminal-Bench | 17.1% | — |
| APEX-Agents | — | 63.3% |
| Berkeley Function Calling Leaderboard | 56.2% | — |
| TheAgentCompany | 41.1% | — |
| τ²-bench Banking | — | 55.1% |
| BALROG | 33.5% | — |
| GDP.pdf | — | 23.2% |
| Vending-Bench 2 | 548.84 | — |
Reasoning Qwen3.8 Max leads
Gemini 2.5 Flash: 18.1 (#286), Qwen3.8 Max: 54.4 (#26)
| Benchmark | Gemini 2.5 Flash | Qwen3.8 Max |
|---|---|---|
| CritPt | 1.1% | 20% |
| LMArena Hard Prompts | 1422 | 1496 |
| DTBench | 76.5% | 92% |
| LMCA | 27.5% | 46.2% |
| Epoch Capabilities Index | 143.03 | 156.41 |
| ARC-AGI-2 | 2.5% | — |
| SimpleBench | 41.2% | — |
| Kagi LLM Benchmark | 56.8% | — |
| NYT Connections (extended) | — | 88.3% |
| ARC-AGI-1 | 33.3% | — |
| Chess Puzzles | — | 40% |
| EnigmaEval | 2.7% | — |
| Mystery Game Puzzles | — | 38% |
| ForecastBench | 60.6 | — |
Math Qwen3.8 Max leads
Gemini 2.5 Flash: 39.9 (#98), Qwen3.8 Max: 73.2 (#20)
| Benchmark | Gemini 2.5 Flash | Qwen3.8 Max |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 73.1% | 100% |
| LMArena Math | 1415 | 1499 |
| FrontierMath (Tiers 1-3) | — | 74.7% |
| FrontierMath Tier 4 | — | 46.3% |
| ProofBench | — | 58% |
| Omni-MATH | 38.5% | — |
| FrontierMath (Feb 2025 set) | 4.8% | — |
| FrontierMath Tier 4 (v1) | 4.2% | — |
Knowledge Qwen3.8 Max leads
Gemini 2.5 Flash: 36.4 (#168), Qwen3.8 Max: 61.7 (#27)
| Benchmark | Gemini 2.5 Flash | Qwen3.8 Max |
|---|---|---|
| LMArena Expert | 1426 | 1507 |
| GPQA Diamond | — | 92.7% |
| Humanity's Last Exam | 12.1% | — |
| SimpleQA Verified | — | 47.3% |
| MMLU-Pro | 63.9% | — |
| Confabulations | 16.8% | — |
| Vectara Hallucination Rate | 7.8% | — |
| GPQA (HELM) | 39% | — |
Multimodal Gemini 2.5 Flash leads
Gemini 2.5 Flash: 41.8 (#32), Qwen3.8 Max: 37.2 (#75)
| Benchmark | Gemini 2.5 Flash | Qwen3.8 Max |
|---|---|---|
| LMArena Vision | 1253 | 1314 |
| GeoBench | 76% | — |
| VPCT | 46.2% | — |
| Furniture Assembly | — | 20% |
| SpatialViz-Bench | 36.9% | — |
Multilingual Qwen3.8 Max leads
Gemini 2.5 Flash: 52.3 (#88), Qwen3.8 Max: 56.7 (#18)
| Benchmark | Gemini 2.5 Flash | Qwen3.8 Max |
|---|---|---|
| LMArena Non-English | 1409 | 1472 |
| LMArena Chinese | 1450 | 1538 |
| LMArena French | 1433 | 1503 |
| LMArena German | 1418 | 1483 |
| LMArena Japanese | 1405 | 1467 |
| LMArena Korean | 1385 | 1461 |
| LMArena Russian | 1415 | 1481 |
| LMArena Spanish | 1421 | 1492 |
Instruction Following Qwen3.8 Max leads
Gemini 2.5 Flash: 75.7 (#54), Qwen3.8 Max: 77.6 (#17)
| Benchmark | Gemini 2.5 Flash | Qwen3.8 Max |
|---|---|---|
| LMArena Instruction Following | 1405 | 1479 |
| IFEval | 89.8% | — |
Long Context Gemini 2.5 Flash leads
Gemini 2.5 Flash: 47.5 (#17), Qwen3.8 Max: 45.6 (#31)
| Benchmark | Gemini 2.5 Flash | Qwen3.8 Max |
|---|---|---|
| LMArena Longer Query | 1419 | 1489 |
| Fiction.LiveBench | 77.8% | — |
Writing & Preference Qwen3.8 Max leads
Gemini 2.5 Flash: 53.8 (#157), Qwen3.8 Max: 67.1 (#30)
| Benchmark | Gemini 2.5 Flash | Qwen3.8 Max |
|---|---|---|
| LMArena Text | 1417 | 1483 |
| LMArena Creative Writing | 1400 | 1479 |
| LMArena Multi-Turn | 1408 | 1489 |
| Short-Story Creative Writing | 76.5% | — |
| EQ-Bench Creative Writing | 1137 | — |
| WildBench | 81.7% | — |
Frequently asked questions
Is Gemini 2.5 Flash better than Qwen3.8 Max?
Qwen3.8 Max is the stronger model overall, scoring 56.8 to 39.3 on the Noometry Index. Gemini 2.5 Flash costs 3.5× less per token, which makes it the better buy when Qwen3.8 Max's lead doesn't matter for your workload.
Which is cheaper, Gemini 2.5 Flash or Qwen3.8 Max?
Gemini 2.5 Flash is cheaper. It lists at $0.30 per million input tokens and $2.50 per million output tokens; Qwen3.8 Max lists at $2 and $6.
Is Gemini 2.5 Flash or Qwen3.8 Max better for coding?
Qwen3.8 Max scores higher on coding benchmarks: 53.5 versus 35.8 in the Noometry coding category.
Which has the bigger context window?
Gemini 2.5 Flash does, with 1.05M tokens against 1M.
How many benchmarks do Gemini 2.5 Flash and Qwen3.8 Max share?
23 benchmarks have published results for both models. Gemini 2.5 Flash has 54 scored results on Noometry and Qwen3.8 Max has 39.