Model comparison
Gemini 3.1 Flash Lite vs Qwen3-30B-A3B
Gemini 3.1 Flash Lite is the stronger model overall, scoring 40.8 to 38.9 on the Noometry Index. Qwen3-30B-A3B costs 2.6× less per token, which makes it the better buy when Gemini 3.1 Flash Lite's lead doesn't matter for your workload.
Last verified . 27 shared benchmarks.
Google
Alibaba (Qwen)Summary
- They share 27 benchmarks with published results for both. Gemini 3.1 Flash Lite scores higher in 9 categories and Qwen3-30B-A3B in 0 categories; 4 gaps are clear of the uncertainty.
- The widest gap is in long context, where Gemini 3.1 Flash Lite leads 42.5 to 31.0.
- The biggest single-benchmark swing is WeirdML: 52.2% for Gemini 3.1 Flash Lite and 29.8% for Qwen3-30B-A3B.
- Qwen3-30B-A3B is cheaper at $0.12 / $0.50 per million input/output tokens, against $0.25 / $1.50 for Gemini 3.1 Flash Lite.
- Gemini 3.1 Flash Lite accepts more context: 1.05M tokens versus 41K.
- Qwen3-30B-A3B has downloadable open weights; the other is API-only.
Side by side
| Gemini 3.1 Flash Lite | Qwen3-30B-A3B | |
|---|---|---|
| Provider | Alibaba (Qwen) | |
| Noometry Index | 40.8 | 38.9 |
| Released | 2026-03-03 | 2025-04-28 |
| Weights | Proprietary | Open |
| Context window | 1.05M | 41K |
| Max output | 66K | 16K |
| Input $ / M tokens | $0.25 | $0.12 |
| Output $ / M tokens | $1.50 | $0.50 |
| Results tracked | 38 | 32 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Too close to call
Gemini 3.1 Flash Lite: 37.8 (#188), Qwen3-30B-A3B: 37.5 (#194)
| Benchmark | Gemini 3.1 Flash Lite | Qwen3-30B-A3B |
|---|---|---|
| SciCode | 41.9% | 33.3% |
| WeirdML | 52.2% | 29.8% |
| LMArena Coding | 1400 | 1416 |
| LMArena WebDev | 1256 | — |
| ALE-Bench | 797.73 | — |
Agentic & Tool Use Too close to call
Gemini 3.1 Flash Lite: 30.2 (#79), Qwen3-30B-A3B: 29.8 (#82)
| Benchmark | Gemini 3.1 Flash Lite | Qwen3-30B-A3B |
|---|---|---|
| Berkeley Function Calling Leaderboard | — | 41.4% |
| DeepResearch Bench | 37.3% | — |
Reasoning Too close to call
Gemini 3.1 Flash Lite: 22.9 (#186), Qwen3-30B-A3B: 22.2 (#204)
| Benchmark | Gemini 3.1 Flash Lite | Qwen3-30B-A3B |
|---|---|---|
| Kagi LLM Benchmark | 67.2% | 54.9% |
| CritPt | 1.1% | 0.3% |
| Chess Puzzles | 25% | 8% |
| LMArena Hard Prompts | 1407 | 1398 |
| DTBench | 76.8% | 69.3% |
| LMCA | 35% | 22.4% |
| Epoch Capabilities Index | 144.47 | 139.63 |
| NYT Connections (extended) | 8.2% | — |
| EnigmaEval | 3% | — |
| Thematic Generalization | 63.3% | — |
| ForecastBench | 54.4 | — |
Math Gemini 3.1 Flash Lite leads
Gemini 3.1 Flash Lite: 40.7 (#90), Qwen3-30B-A3B: 37.4 (#157)
| Benchmark | Gemini 3.1 Flash Lite | Qwen3-30B-A3B |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 80% | 70.3% |
| LMArena Math | 1428 | 1394 |
| FrontierMath (Tiers 1-3) | 27.7% | — |
| MathArena Final-Answer Competitions | — | 47.8% |
Knowledge Too close to call
Gemini 3.1 Flash Lite: 41.9 (#104), Qwen3-30B-A3B: 41.8 (#105)
| Benchmark | Gemini 3.1 Flash Lite | Qwen3-30B-A3B |
|---|---|---|
| GPQA Diamond | 81.8% | 70.1% |
| LMArena Expert | 1398 | 1396 |
| Humanity's Last Exam | 8.6% | — |
| Confabulations | — | 12.3% |
| Vectara Hallucination Rate | 8.2% | — |
Multimodal Not comparable
Gemini 3.1 Flash Lite: 39.4 (#60), Qwen3-30B-A3B: —
| Benchmark | Gemini 3.1 Flash Lite | Qwen3-30B-A3B |
|---|---|---|
| LMArena Vision | 1240 | — |
Multilingual Gemini 3.1 Flash Lite leads
Gemini 3.1 Flash Lite: 52.3 (#86), Qwen3-30B-A3B: 49.5 (#132)
| Benchmark | Gemini 3.1 Flash Lite | Qwen3-30B-A3B |
|---|---|---|
| LMArena Non-English | 1411 | 1372 |
| LMArena Chinese | 1461 | 1433 |
| LMArena French | 1424 | 1418 |
| LMArena German | 1429 | 1380 |
| LMArena Japanese | 1413 | 1337 |
| LMArena Korean | 1392 | 1331 |
| LMArena Russian | 1420 | 1370 |
| LMArena Spanish | 1421 | 1404 |
Instruction Following Too close to call
Gemini 3.1 Flash Lite: 72.7 (#131), Qwen3-30B-A3B: 72.0 (#142)
| Benchmark | Gemini 3.1 Flash Lite | Qwen3-30B-A3B |
|---|---|---|
| LMArena Instruction Following | 1377 | 1363 |
Long Context Gemini 3.1 Flash Lite leads
Gemini 3.1 Flash Lite: 42.5 (#122), Qwen3-30B-A3B: 31.0 (#283)
| Benchmark | Gemini 3.1 Flash Lite | Qwen3-30B-A3B |
|---|---|---|
| LMArena Longer Query | 1394 | 1379 |
| Fiction.LiveBench | — | 40.6% |
Writing & Preference Gemini 3.1 Flash Lite leads
Gemini 3.1 Flash Lite: 60.9 (#94), Qwen3-30B-A3B: 55.6 (#143)
| Benchmark | Gemini 3.1 Flash Lite | Qwen3-30B-A3B |
|---|---|---|
| LMArena Text | 1416 | 1384 |
| LMArena Creative Writing | 1401 | 1317 |
| LMArena Multi-Turn | 1417 | 1378 |
| Short-Story Creative Writing | — | 75.3% |
Frequently asked questions
Is Gemini 3.1 Flash Lite better than Qwen3-30B-A3B?
Gemini 3.1 Flash Lite is the stronger model overall, scoring 40.8 to 38.9 on the Noometry Index. Qwen3-30B-A3B costs 2.6× less per token, which makes it the better buy when Gemini 3.1 Flash Lite's lead doesn't matter for your workload.
Which is cheaper, Gemini 3.1 Flash Lite or Qwen3-30B-A3B?
Qwen3-30B-A3B is cheaper. It lists at $0.12 per million input tokens and $0.50 per million output tokens; Gemini 3.1 Flash Lite lists at $0.25 and $1.50.
Is Gemini 3.1 Flash Lite or Qwen3-30B-A3B better for coding?
They score almost the same on coding (37.8 vs 37.5); test both on your own repository before choosing.
Which has the bigger context window?
Gemini 3.1 Flash Lite does, with 1.05M tokens against 41K.
How many benchmarks do Gemini 3.1 Flash Lite and Qwen3-30B-A3B share?
27 benchmarks have published results for both models. Gemini 3.1 Flash Lite has 38 scored results on Noometry and Qwen3-30B-A3B has 32.