Model comparison
Gemini 3 Flash Preview vs Qwen3.8 Max
Qwen3.8 Max is the stronger model overall, scoring 56.8 to 52.3 on the Noometry Index. Gemini 3 Flash Preview costs 2.7× less per token, which makes it the better buy when Qwen3.8 Max's lead doesn't matter for your workload.
Last verified . 33 shared benchmarks.
Google
Alibaba (Qwen)Summary
- They share 33 benchmarks with published results for both. Gemini 3 Flash Preview scores higher in 1 category and Qwen3.8 Max in 9 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in math, where Qwen3.8 Max leads 73.2 to 51.7.
- The biggest single-benchmark swing is ProofBench: 15% for Gemini 3 Flash Preview and 58% for Qwen3.8 Max.
- Gemini 3 Flash Preview is cheaper at $0.50 / $3 per million input/output tokens, against $2 / $6 for Qwen3.8 Max.
- Gemini 3 Flash Preview accepts more context: 1.05M tokens versus 1M.
Side by side
| Gemini 3 Flash Preview | Qwen3.8 Max | |
|---|---|---|
| Provider | Alibaba (Qwen) | |
| Noometry Index | 52.3 | 56.8 |
| Released | 2025-12-17 | 2026-08-02 |
| Weights | Proprietary | Proprietary |
| Context window | 1.05M | 1M |
| Max output | 66K | 131K |
| Input $ / M tokens | $0.50 | $2 |
| Output $ / M tokens | $3 | $6 |
| Results tracked | 59 | 39 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Qwen3.8 Max leads
Gemini 3 Flash Preview: 50.9 (#42), Qwen3.8 Max: 53.5 (#29)
| Benchmark | Gemini 3 Flash Preview | Qwen3.8 Max |
|---|---|---|
| LMArena WebDev | 1439 | 1674 |
| LMArena Coding | 1460 | 1502 |
| SWE-bench Verified | 75.4% | — |
| DeepSWE | — | 57.5% |
| SWE-bench Verified (bash only) | 75.8% | — |
| SWE-bench Multilingual | 72.7% | — |
| FrontierSWE | — | 17.8% |
| SciCode | — | 53.2% |
| GSO | 9.8% | — |
| WeirdML | 61.6% | — |
| ALE-Bench | 1,367 | — |
Agentic & Tool Use Qwen3.8 Max leads
Gemini 3 Flash Preview: 38.7 (#29), Qwen3.8 Max: 45.4 (#14)
| Benchmark | Gemini 3 Flash Preview | Qwen3.8 Max |
|---|---|---|
| τ²-bench Banking | 27.3% | 55.1% |
| GDP.pdf | 10% | 23.2% |
| Terminal-Bench | 64.3% | — |
| APEX-Agents | — | 63.3% |
| τ²-bench Airline | 82.5% | — |
| τ²-bench Retail | 76.8% | — |
| τ²-bench Telecom | 91.2% | — |
| DeepResearch Bench | 49.8% | — |
| BALROG | 48.1% | — |
| LMArena Search | 1198 | — |
| Vending-Bench 2 | 3,635 | — |
Reasoning Qwen3.8 Max leads
Gemini 3 Flash Preview: 49.2 (#37), Qwen3.8 Max: 54.4 (#26)
| Benchmark | Gemini 3 Flash Preview | Qwen3.8 Max |
|---|---|---|
| NYT Connections (extended) | 83.1% | 88.3% |
| Chess Puzzles | 40% | 40% |
| LMArena Hard Prompts | 1465 | 1496 |
| Mystery Game Puzzles | 26% | 38% |
| DTBench | 89.1% | 92% |
| LMCA | 43.1% | 46.2% |
| Epoch Capabilities Index | 151.8 | 156.41 |
| ARC-AGI-2 | 33.6% | — |
| SimpleBench | 61.1% | — |
| ARC-AGI-1 | 84.7% | — |
| CritPt | — | 20% |
| ForecastBench | 58.5 | — |
Math Qwen3.8 Max leads
Gemini 3 Flash Preview: 51.7 (#55), Qwen3.8 Max: 73.2 (#20)
| Benchmark | Gemini 3 Flash Preview | Qwen3.8 Max |
|---|---|---|
| FrontierMath (Tiers 1-3) | 51.2% | 74.7% |
| FrontierMath Tier 4 | 17.1% | 46.3% |
| OTIS Mock AIME 2024-2025 | 95.6% | 100% |
| ProofBench | 15% | 58% |
| LMArena Math | 1473 | 1499 |
| MathArena Final-Answer Competitions | 67.6% | — |
| FrontierMath (Feb 2025 set) | 35.6% | — |
| FrontierMath Tier 4 (v1) | 4.2% | — |
Knowledge Qwen3.8 Max leads
Gemini 3 Flash Preview: 58.8 (#33), Qwen3.8 Max: 61.7 (#27)
| Benchmark | Gemini 3 Flash Preview | Qwen3.8 Max |
|---|---|---|
| GPQA Diamond | 89.4% | 92.7% |
| SimpleQA Verified | 66.8% | 47.3% |
| LMArena Expert | 1462 | 1507 |
| Vectara Hallucination Rate | 13.5% | — |
Multimodal Gemini 3 Flash Preview leads
Gemini 3 Flash Preview: 45.5 (#16), Qwen3.8 Max: 37.2 (#75)
| Benchmark | Gemini 3 Flash Preview | Qwen3.8 Max |
|---|---|---|
| LMArena Vision | 1285 | 1314 |
| GeoBench | 88% | — |
| VPCT | 72.6% | — |
| Blueprint-Bench 2 | 0% | — |
| Furniture Assembly | — | 20% |
| LMArena Document | 1413 | — |
Multilingual Too close to call
Gemini 3 Flash Preview: 55.7 (#27), Qwen3.8 Max: 56.7 (#18)
| Benchmark | Gemini 3 Flash Preview | Qwen3.8 Max |
|---|---|---|
| LMArena Non-English | 1458 | 1472 |
| LMArena Chinese | 1511 | 1538 |
| LMArena French | 1477 | 1503 |
| LMArena German | 1497 | 1483 |
| LMArena Japanese | 1489 | 1467 |
| LMArena Korean | 1443 | 1461 |
| LMArena Russian | 1480 | 1481 |
| LMArena Spanish | 1469 | 1492 |
Instruction Following Qwen3.8 Max leads
Gemini 3 Flash Preview: 75.7 (#56), Qwen3.8 Max: 77.6 (#17)
| Benchmark | Gemini 3 Flash Preview | Qwen3.8 Max |
|---|---|---|
| LMArena Instruction Following | 1437 | 1479 |
Long Context Qwen3.8 Max leads
Gemini 3 Flash Preview: 44.4 (#67), Qwen3.8 Max: 45.6 (#31)
| Benchmark | Gemini 3 Flash Preview | Qwen3.8 Max |
|---|---|---|
| LMArena Longer Query | 1452 | 1489 |
Writing & Preference Qwen3.8 Max leads
Gemini 3 Flash Preview: 65.5 (#45), Qwen3.8 Max: 67.1 (#30)
| Benchmark | Gemini 3 Flash Preview | Qwen3.8 Max |
|---|---|---|
| LMArena Text | 1466 | 1483 |
| LMArena Creative Writing | 1457 | 1479 |
| LMArena Multi-Turn | 1471 | 1489 |
Frequently asked questions
Is Gemini 3 Flash Preview better than Qwen3.8 Max?
Qwen3.8 Max is the stronger model overall, scoring 56.8 to 52.3 on the Noometry Index. Gemini 3 Flash Preview costs 2.7× 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 3 Flash Preview or Qwen3.8 Max?
Gemini 3 Flash Preview is cheaper. It lists at $0.50 per million input tokens and $3 per million output tokens; Qwen3.8 Max lists at $2 and $6.
Is Gemini 3 Flash Preview or Qwen3.8 Max better for coding?
Qwen3.8 Max scores higher on coding benchmarks: 53.5 versus 50.9 in the Noometry coding category.
Which has the bigger context window?
Gemini 3 Flash Preview does, with 1.05M tokens against 1M.
How many benchmarks do Gemini 3 Flash Preview and Qwen3.8 Max share?
33 benchmarks have published results for both models. Gemini 3 Flash Preview has 59 scored results on Noometry and Qwen3.8 Max has 39.