Model comparison
Gemini 3.1 Flash Lite vs Qwen3-Coder 480B-A35B Instruct
Gemini 3.1 Flash Lite is the stronger model overall, scoring 40.4 to 37.7 on the Noometry Index.
Last verified . 21 shared benchmarks.
Google
Alibaba (Qwen)Summary
- They share 21 benchmarks with published results for both. Gemini 3.1 Flash Lite scores higher in 8 categories and Qwen3-Coder 480B-A35B Instruct in 1 category; 8 gaps are clear of the uncertainty.
- The widest gap is in agentic & tool use, where Gemini 3.1 Flash Lite leads 29.9 to 23.7.
- The biggest single-benchmark swing is Kagi LLM Benchmark: 67.2% for Gemini 3.1 Flash Lite and 49.5% for Qwen3-Coder 480B-A35B Instruct.
- Gemini 3.1 Flash Lite is cheaper at $0.25 / $1.50 per million input/output tokens, against $1.50 / $7.50 for Qwen3-Coder 480B-A35B Instruct.
- Gemini 3.1 Flash Lite accepts more context: 1.05M tokens versus 262K.
- Qwen3-Coder 480B-A35B Instruct has downloadable open weights; the other is API-only.
Side by side
| Gemini 3.1 Flash Lite | Qwen3-Coder 480B-A35B Instruct | |
|---|---|---|
| Provider | Alibaba (Qwen) | |
| Noometry Index | 40.4 | 37.7 |
| Released | 2026-03-03 | 2025-04 |
| Weights | Proprietary | Open |
| Context window | 1.05M | 262K |
| Max output | 66K | 66K |
| Input $ / M tokens | $0.25 | $1.50 |
| Output $ / M tokens | $1.50 | $7.50 |
| Results tracked | 38 | 25 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Gemini 3.1 Flash Lite leads
Gemini 3.1 Flash Lite: 37.5 (#179), Qwen3-Coder 480B-A35B Instruct: 36.2 (#199)
| Benchmark | Gemini 3.1 Flash Lite | Qwen3-Coder 480B-A35B Instruct |
|---|---|---|
| LMArena WebDev | 1256 | 1275 |
| WeirdML | 52.2% | 41.2% |
| LMArena Coding | 1400 | 1413 |
| ALE-Bench | 797.73 | 461.45 |
| SWE-bench Verified (bash only) | — | 55.4% |
| SciCode | 41.9% | — |
| GSO | — | 4.9% |
| AlgoTune | — | 1.44 |
Agentic & Tool Use Gemini 3.1 Flash Lite leads
Gemini 3.1 Flash Lite: 29.9 (#72), Qwen3-Coder 480B-A35B Instruct: 23.7 (#116)
| Benchmark | Gemini 3.1 Flash Lite | Qwen3-Coder 480B-A35B Instruct |
|---|---|---|
| Terminal-Bench | — | 27.2% |
| DeepResearch Bench | 37.3% | — |
Reasoning Qwen3-Coder 480B-A35B Instruct leads
Gemini 3.1 Flash Lite: 22.3 (#181), Qwen3-Coder 480B-A35B Instruct: 24.7 (#141)
| Benchmark | Gemini 3.1 Flash Lite | Qwen3-Coder 480B-A35B Instruct |
|---|---|---|
| Kagi LLM Benchmark | 67.2% | 49.5% |
| LMArena Hard Prompts | 1407 | 1372 |
| NYT Connections (extended) | 8.2% | — |
| CritPt | 1.1% | — |
| Chess Puzzles | 25% | — |
| EnigmaEval | 3% | — |
| Thematic Generalization | 63.3% | — |
| DTBench | 76.8% | — |
| LMCA | 35% | — |
| Epoch Capabilities Index | 144.45 | — |
| ForecastBench | 54.4 | — |
Math Gemini 3.1 Flash Lite leads
Gemini 3.1 Flash Lite: 39.5 (#82), Qwen3-Coder 480B-A35B Instruct: 34.9 (#151)
| Benchmark | Gemini 3.1 Flash Lite | Qwen3-Coder 480B-A35B Instruct |
|---|---|---|
| LMArena Math | 1427 | 1364 |
| FrontierMath (Tiers 1-3) | 27.7% | — |
| OTIS Mock AIME 2024-2025 | 80% | — |
Knowledge Gemini 3.1 Flash Lite leads
Gemini 3.1 Flash Lite: 41.7 (#95), Qwen3-Coder 480B-A35B Instruct: 36.4 (#159)
| Benchmark | Gemini 3.1 Flash Lite | Qwen3-Coder 480B-A35B Instruct |
|---|---|---|
| LMArena Expert | 1397 | 1337 |
| GPQA Diamond | 81.8% | — |
| Humanity's Last Exam | 8.6% | — |
| Vectara Hallucination Rate | 8.2% | — |
Multimodal Not comparable
Gemini 3.1 Flash Lite: 38.7 (#57), Qwen3-Coder 480B-A35B Instruct: —
| Benchmark | Gemini 3.1 Flash Lite | Qwen3-Coder 480B-A35B Instruct |
|---|---|---|
| LMArena Vision | 1240 | — |
Multilingual Gemini 3.1 Flash Lite leads
Gemini 3.1 Flash Lite: 52.5 (#82), Qwen3-Coder 480B-A35B Instruct: 47.8 (#143)
| Benchmark | Gemini 3.1 Flash Lite | Qwen3-Coder 480B-A35B Instruct |
|---|---|---|
| LMArena Non-English | 1410 | 1345 |
| LMArena Chinese | 1461 | 1356 |
| LMArena French | 1423 | 1394 |
| LMArena German | 1429 | 1324 |
| LMArena Japanese | 1411 | 1309 |
| LMArena Korean | 1393 | 1307 |
| LMArena Russian | 1420 | 1367 |
| LMArena Spanish | 1420 | 1360 |
Instruction Following Gemini 3.1 Flash Lite leads
Gemini 3.1 Flash Lite: 72.9 (#126), Qwen3-Coder 480B-A35B Instruct: 71.8 (#142)
| Benchmark | Gemini 3.1 Flash Lite | Qwen3-Coder 480B-A35B Instruct |
|---|---|---|
| LMArena Instruction Following | 1376 | 1356 |
Long Context Too close to call
Gemini 3.1 Flash Lite: 42.5 (#117), Qwen3-Coder 480B-A35B Instruct: 42.0 (#126)
| Benchmark | Gemini 3.1 Flash Lite | Qwen3-Coder 480B-A35B Instruct |
|---|---|---|
| LMArena Longer Query | 1394 | 1379 |
Writing & Preference Gemini 3.1 Flash Lite leads
Gemini 3.1 Flash Lite: 61.0 (#90), Qwen3-Coder 480B-A35B Instruct: 55.5 (#142)
| Benchmark | Gemini 3.1 Flash Lite | Qwen3-Coder 480B-A35B Instruct |
|---|---|---|
| LMArena Text | 1416 | 1356 |
| LMArena Creative Writing | 1401 | 1333 |
| LMArena Multi-Turn | 1417 | 1365 |
Frequently asked questions
Is Gemini 3.1 Flash Lite better than Qwen3-Coder 480B-A35B Instruct?
Gemini 3.1 Flash Lite is the stronger model overall, scoring 40.4 to 37.7 on the Noometry Index.
Which is cheaper, Gemini 3.1 Flash Lite or Qwen3-Coder 480B-A35B Instruct?
Gemini 3.1 Flash Lite is cheaper. It lists at $0.25 per million input tokens and $1.50 per million output tokens; Qwen3-Coder 480B-A35B Instruct lists at $1.50 and $7.50.
Is Gemini 3.1 Flash Lite or Qwen3-Coder 480B-A35B Instruct better for coding?
Gemini 3.1 Flash Lite scores higher on coding benchmarks: 37.5 versus 36.2 in the Noometry coding category.
Which has the bigger context window?
Gemini 3.1 Flash Lite does, with 1.05M tokens against 262K.
How many benchmarks do Gemini 3.1 Flash Lite and Qwen3-Coder 480B-A35B Instruct share?
21 benchmarks have published results for both models. Gemini 3.1 Flash Lite has 38 scored results on Noometry and Qwen3-Coder 480B-A35B Instruct has 25.