Model comparison
Gemini 3.5 Flash Lite vs Gemma 4 31B IT
Gemma 4 31B IT is the stronger model overall, scoring 43.5 to 41.5 on the Noometry Index.
Last verified . 28 shared benchmarks.
GoogleSummary
- They share 28 benchmarks with published results for both. Gemini 3.5 Flash Lite scores higher in 4 categories and Gemma 4 31B IT in 5 categories; 3 gaps are clear of the uncertainty.
- The widest gap is in math, where Gemma 4 31B IT leads 43.2 to 25.9.
- The biggest single-benchmark swing is Chess Puzzles: 22% for Gemini 3.5 Flash Lite and 5% for Gemma 4 31B IT.
- Gemma 4 31B IT is cheaper at $0.09 / $0.34 per million input/output tokens, against $0.30 / $2.50 for Gemini 3.5 Flash Lite.
- Gemini 3.5 Flash Lite accepts more context: 1.05M tokens versus 262K.
- Gemma 4 31B IT has downloadable open weights; the other is API-only.
Side by side
| Gemini 3.5 Flash Lite | Gemma 4 31B IT | |
|---|---|---|
| Provider | ||
| Noometry Index | 41.5 | 43.5 |
| Released | 2026-07-21 | 2026-04-02 |
| Weights | Proprietary | Open |
| Context window | 1.05M | 262K |
| Max output | 66K | 33K |
| Input $ / M tokens | $0.30 | $0.09 |
| Output $ / M tokens | $2.50 | $0.34 |
| Results tracked | 39 | 35 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Too close to call
Gemini 3.5 Flash Lite: 42.4 (#104), Gemma 4 31B IT: 42.3 (#108)
| Benchmark | Gemini 3.5 Flash Lite | Gemma 4 31B IT |
|---|---|---|
| LMArena WebDev | 1440 | 1366 |
| SciCode | 41.3% | 43.4% |
| WeirdML | 39% | 52.3% |
| LMArena Coding | 1453 | 1459 |
| ALE-Bench | 765.27 | 925.5 |
Agentic & Tool Use Not comparable
Gemini 3.5 Flash Lite: 25.7 (#107), Gemma 4 31B IT: —
| Benchmark | Gemini 3.5 Flash Lite | Gemma 4 31B IT |
|---|---|---|
| APEX-Agents | 29.3% | — |
| GDP.pdf | 10% | — |
Reasoning Too close to call
Gemini 3.5 Flash Lite: 27.8 (#114), Gemma 4 31B IT: 27.2 (#122)
| Benchmark | Gemini 3.5 Flash Lite | Gemma 4 31B IT |
|---|---|---|
| NYT Connections (extended) | 60.4% | 70.6% |
| CritPt | 0% | 1.4% |
| Chess Puzzles | 22% | 5% |
| LMArena Hard Prompts | 1441 | 1448 |
| DTBench | 83.5% | 82.7% |
| LMCA | 37.6% | 39.3% |
| Epoch Capabilities Index | 145.13 | 142.74 |
| ARC-AGI-2 | 10.3% | — |
| Kagi LLM Benchmark | — | 63.5% |
| ARC-AGI-1 | 53.5% | — |
| Thematic Generalization | — | 53% |
| Mystery Game Puzzles | 19% | — |
| Surface Evolver Bench | — | 30.6% |
Math Gemma 4 31B IT leads
Gemini 3.5 Flash Lite: 25.9 (#264), Gemma 4 31B IT: 43.2 (#81)
| Benchmark | Gemini 3.5 Flash Lite | Gemma 4 31B IT |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 71.1% | 73.3% |
| LMArena Math | 1433 | 1465 |
| FrontierMath (Tiers 1-3) | 26% | — |
| FrontierMath Tier 4 | 0% | — |
| ProofBench | 13% | — |
Knowledge Gemini 3.5 Flash Lite leads
Gemini 3.5 Flash Lite: 50.1 (#72), Gemma 4 31B IT: 37.9 (#151)
| Benchmark | Gemini 3.5 Flash Lite | Gemma 4 31B IT |
|---|---|---|
| GPQA Diamond | 83.3% | 75.8% |
| LMArena Expert | 1435 | 1465 |
| SimpleQA Verified | — | 10.4% |
| Vectara Hallucination Rate | — | 7.4% |
Multimodal Too close to call
Gemini 3.5 Flash Lite: 41.1 (#40), Gemma 4 31B IT: 41.6 (#34)
| Benchmark | Gemini 3.5 Flash Lite | Gemma 4 31B IT |
|---|---|---|
| LMArena Vision | 1268 | 1277 |
| LMArena Document | — | 1425 |
Multilingual Too close to call
Gemini 3.5 Flash Lite: 53.5 (#63), Gemma 4 31B IT: 53.8 (#57)
| Benchmark | Gemini 3.5 Flash Lite | Gemma 4 31B IT |
|---|---|---|
| LMArena Non-English | 1427 | 1431 |
| LMArena Chinese | 1469 | 1476 |
| LMArena French | 1451 | 1435 |
| LMArena Russian | 1443 | 1460 |
| LMArena Spanish | 1442 | 1444 |
| LMArena German | 1454 | — |
| LMArena Japanese | 1427 | — |
| LMArena Korean | 1407 | — |
Instruction Following Too close to call
Gemini 3.5 Flash Lite: 74.8 (#85), Gemma 4 31B IT: 75.5 (#61)
| Benchmark | Gemini 3.5 Flash Lite | Gemma 4 31B IT |
|---|---|---|
| LMArena Instruction Following | 1420 | 1433 |
Long Context Too close to call
Gemini 3.5 Flash Lite: 43.9 (#84), Gemma 4 31B IT: 44.2 (#71)
| Benchmark | Gemini 3.5 Flash Lite | Gemma 4 31B IT |
|---|---|---|
| LMArena Longer Query | 1435 | 1446 |
Writing & Preference Gemini 3.5 Flash Lite leads
Gemini 3.5 Flash Lite: 64.2 (#57), Gemma 4 31B IT: 60.5 (#96)
| Benchmark | Gemini 3.5 Flash Lite | Gemma 4 31B IT |
|---|---|---|
| LMArena Text | 1435 | 1443 |
| LMArena Creative Writing | 1420 | 1415 |
| EQ-Bench Creative Writing | 1559 | 1368 |
| LMArena Multi-Turn | 1445 | 1452 |
| EQ-Bench 4 | — | 1120 |
Frequently asked questions
Is Gemini 3.5 Flash Lite better than Gemma 4 31B IT?
Gemma 4 31B IT is the stronger model overall, scoring 43.5 to 41.5 on the Noometry Index.
Which is cheaper, Gemini 3.5 Flash Lite or Gemma 4 31B IT?
Gemma 4 31B IT is cheaper. It lists at $0.09 per million input tokens and $0.34 per million output tokens; Gemini 3.5 Flash Lite lists at $0.30 and $2.50.
Is Gemini 3.5 Flash Lite or Gemma 4 31B IT better for coding?
They score almost the same on coding (42.4 vs 42.3); test both on your own repository before choosing.
Which has the bigger context window?
Gemini 3.5 Flash Lite does, with 1.05M tokens against 262K.
How many benchmarks do Gemini 3.5 Flash Lite and Gemma 4 31B IT share?
28 benchmarks have published results for both models. Gemini 3.5 Flash Lite has 39 scored results on Noometry and Gemma 4 31B IT has 35.