Model comparison
Gemini 3.1 Flash Lite vs MiniMax-M3
MiniMax-M3 is the stronger model overall, scoring 43.8 to 40.8 on the Noometry Index.
Last verified . 30 shared benchmarks.
Google
MiniMaxSummary
- They share 30 benchmarks with published results for both. Gemini 3.1 Flash Lite scores higher in 2 categories and MiniMax-M3 in 8 categories; 7 gaps are clear of the uncertainty.
- The widest gap is in knowledge, where MiniMax-M3 leads 58.4 to 41.9.
- The biggest single-benchmark swing is NYT Connections (extended): 8.2% for Gemini 3.1 Flash Lite and 65.1% for MiniMax-M3.
- MiniMax-M3 is cheaper at $0.30 / $1.20 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 1M.
- MiniMax-M3 has downloadable open weights; the other is API-only.
Side by side
| Gemini 3.1 Flash Lite | MiniMax-M3 | |
|---|---|---|
| Provider | MiniMax | |
| Noometry Index | 40.8 | 43.8 |
| Released | 2026-03-03 | 2026-06-01 |
| Weights | Proprietary | Open |
| Context window | 1.05M | 1M |
| Max output | 66K | 512K |
| Input $ / M tokens | $0.25 | $0.30 |
| Output $ / M tokens | $1.50 | $1.20 |
| Results tracked | 38 | 41 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding MiniMax-M3 leads
Gemini 3.1 Flash Lite: 37.8 (#188), MiniMax-M3: 41.8 (#118)
| Benchmark | Gemini 3.1 Flash Lite | MiniMax-M3 |
|---|---|---|
| LMArena WebDev | 1256 | 1482 |
| SciCode | 41.9% | 47.1% |
| LMArena Coding | 1400 | 1469 |
| ALE-Bench | 797.73 | 640.02 |
| FrontierCode | — | 14.7% |
| WeirdML | 52.2% | — |
Agentic & Tool Use Gemini 3.1 Flash Lite leads
Gemini 3.1 Flash Lite: 30.2 (#79), MiniMax-M3: 22.6 (#130)
| Benchmark | Gemini 3.1 Flash Lite | MiniMax-M3 |
|---|---|---|
| APEX-Agents | — | 37.7% |
| OSWorld 2.0 | — | 4.6% |
| DeepResearch Bench | 37.3% | — |
| GBAEval | — | 0.9% |
| Vending-Bench 2 | — | 2,158 |
Reasoning MiniMax-M3 leads
Gemini 3.1 Flash Lite: 22.9 (#186), MiniMax-M3: 30.1 (#87)
| Benchmark | Gemini 3.1 Flash Lite | MiniMax-M3 |
|---|---|---|
| NYT Connections (extended) | 8.2% | 65.1% |
| CritPt | 1.1% | 3.7% |
| Chess Puzzles | 25% | 14% |
| LMArena Hard Prompts | 1407 | 1447 |
| DTBench | 76.8% | 78.9% |
| LMCA | 35% | 33.7% |
| Epoch Capabilities Index | 144.47 | 146.95 |
| ForecastBench | 54.4 | 61.4 |
| SimpleBench | — | 45.8% |
| Kagi LLM Benchmark | 67.2% | — |
| EnigmaEval | 3% | — |
| Thematic Generalization | 63.3% | — |
| Mystery Game Puzzles | — | 8% |
| Surface Evolver Bench | — | 55% |
Math Too close to call
Gemini 3.1 Flash Lite: 40.7 (#90), MiniMax-M3: 40.0 (#95)
| Benchmark | Gemini 3.1 Flash Lite | MiniMax-M3 |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 80% | 71.1% |
| LMArena Math | 1428 | 1429 |
| FrontierMath (Tiers 1-3) | 27.7% | — |
| ProofBench | — | 18% |
Knowledge MiniMax-M3 leads
Gemini 3.1 Flash Lite: 41.9 (#104), MiniMax-M3: 58.4 (#35)
| Benchmark | Gemini 3.1 Flash Lite | MiniMax-M3 |
|---|---|---|
| GPQA Diamond | 81.8% | 90.9% |
| LMArena Expert | 1398 | 1461 |
| Humanity's Last Exam | 8.6% | — |
| Vectara Hallucination Rate | 8.2% | — |
Multimodal Too close to call
Gemini 3.1 Flash Lite: 39.4 (#60), MiniMax-M3: 40.2 (#51)
| Benchmark | Gemini 3.1 Flash Lite | MiniMax-M3 |
|---|---|---|
| LMArena Vision | 1240 | 1253 |
| LMArena Document | — | 1435 |
Multilingual Too close to call
Gemini 3.1 Flash Lite: 52.3 (#86), MiniMax-M3: 53.0 (#75)
| Benchmark | Gemini 3.1 Flash Lite | MiniMax-M3 |
|---|---|---|
| LMArena Non-English | 1411 | 1420 |
| LMArena Chinese | 1461 | 1463 |
| LMArena French | 1424 | 1447 |
| LMArena German | 1429 | 1426 |
| LMArena Japanese | 1413 | 1381 |
| LMArena Korean | 1392 | 1372 |
| LMArena Russian | 1420 | 1428 |
| LMArena Spanish | 1421 | 1432 |
Instruction Following MiniMax-M3 leads
Gemini 3.1 Flash Lite: 72.7 (#131), MiniMax-M3: 75.5 (#62)
| Benchmark | Gemini 3.1 Flash Lite | MiniMax-M3 |
|---|---|---|
| LMArena Instruction Following | 1377 | 1433 |
Long Context MiniMax-M3 leads
Gemini 3.1 Flash Lite: 42.5 (#122), MiniMax-M3: 44.2 (#72)
| Benchmark | Gemini 3.1 Flash Lite | MiniMax-M3 |
|---|---|---|
| LMArena Longer Query | 1394 | 1445 |
Writing & Preference MiniMax-M3 leads
Gemini 3.1 Flash Lite: 60.9 (#94), MiniMax-M3: 62.1 (#83)
| Benchmark | Gemini 3.1 Flash Lite | MiniMax-M3 |
|---|---|---|
| LMArena Text | 1416 | 1433 |
| LMArena Creative Writing | 1401 | 1404 |
| LMArena Multi-Turn | 1417 | 1442 |
| EQ-Bench 4 | — | 1150 |
Frequently asked questions
Is Gemini 3.1 Flash Lite better than MiniMax-M3?
MiniMax-M3 is the stronger model overall, scoring 43.8 to 40.8 on the Noometry Index.
Which is cheaper, Gemini 3.1 Flash Lite or MiniMax-M3?
MiniMax-M3 is cheaper. It lists at $0.30 per million input tokens and $1.20 per million output tokens; Gemini 3.1 Flash Lite lists at $0.25 and $1.50.
Is Gemini 3.1 Flash Lite or MiniMax-M3 better for coding?
MiniMax-M3 scores higher on coding benchmarks: 41.8 versus 37.8 in the Noometry coding category.
Which has the bigger context window?
Gemini 3.1 Flash Lite does, with 1.05M tokens against 1M.
How many benchmarks do Gemini 3.1 Flash Lite and MiniMax-M3 share?
30 benchmarks have published results for both models. Gemini 3.1 Flash Lite has 38 scored results on Noometry and MiniMax-M3 has 41.