Model comparison
Gemini 2.5 Flash vs MiniMax-M3
MiniMax-M3 is the stronger model overall, scoring 43.8 to 39.3 on the Noometry Index.
Last verified . 27 shared benchmarks.
Google
MiniMaxSummary
- They share 27 benchmarks with published results for both. Gemini 2.5 Flash scores higher in 4 categories and MiniMax-M3 in 6 categories; 7 gaps are clear of the uncertainty.
- The widest gap is in knowledge, where MiniMax-M3 leads 58.4 to 36.4.
- The biggest single-benchmark swing is LMCA: 27.5% for Gemini 2.5 Flash and 33.7% for MiniMax-M3.
- MiniMax-M3 is cheaper at $0.30 / $1.20 per million input/output tokens, against $0.30 / $2.50 for Gemini 2.5 Flash.
- Gemini 2.5 Flash accepts more context: 1.05M tokens versus 1M.
- MiniMax-M3 has downloadable open weights; the other is API-only.
Side by side
| Gemini 2.5 Flash | MiniMax-M3 | |
|---|---|---|
| Provider | MiniMax | |
| Noometry Index | 39.3 | 43.8 |
| Released | 2025-04-17 | 2026-06-01 |
| Weights | Proprietary | Open |
| Context window | 1.05M | 1M |
| Max output | 66K | 512K |
| Input $ / M tokens | $0.30 | $0.30 |
| Output $ / M tokens | $2.50 | $1.20 |
| Results tracked | 54 | 41 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding MiniMax-M3 leads
Gemini 2.5 Flash: 35.8 (#220), MiniMax-M3: 41.8 (#118)
| Benchmark | Gemini 2.5 Flash | MiniMax-M3 |
|---|---|---|
| LMArena Coding | 1424 | 1469 |
| ALE-Bench | 661.88 | 640.02 |
| FrontierCode | — | 14.7% |
| SWE-bench Verified (bash only) | 28.7% | — |
| Aider Polyglot | 55.1% | — |
| LMArena WebDev | — | 1482 |
| SciCode | — | 47.1% |
| WeirdML | 41.9% | — |
Agentic & Tool Use Gemini 2.5 Flash leads
Gemini 2.5 Flash: 30.8 (#74), MiniMax-M3: 22.6 (#130)
| Benchmark | Gemini 2.5 Flash | MiniMax-M3 |
|---|---|---|
| Vending-Bench 2 | 548.84 | 2,158 |
| Terminal-Bench | 17.1% | — |
| APEX-Agents | — | 37.7% |
| Berkeley Function Calling Leaderboard | 56.2% | — |
| OSWorld 2.0 | — | 4.6% |
| TheAgentCompany | 41.1% | — |
| BALROG | 33.5% | — |
| GBAEval | — | 0.9% |
Reasoning MiniMax-M3 leads
Gemini 2.5 Flash: 18.1 (#286), MiniMax-M3: 30.1 (#87)
| Benchmark | Gemini 2.5 Flash | MiniMax-M3 |
|---|---|---|
| SimpleBench | 41.2% | 45.8% |
| CritPt | 1.1% | 3.7% |
| LMArena Hard Prompts | 1422 | 1447 |
| DTBench | 76.5% | 78.9% |
| LMCA | 27.5% | 33.7% |
| Epoch Capabilities Index | 143.03 | 146.95 |
| ForecastBench | 60.6 | 61.4 |
| ARC-AGI-2 | 2.5% | — |
| Kagi LLM Benchmark | 56.8% | — |
| NYT Connections (extended) | — | 65.1% |
| ARC-AGI-1 | 33.3% | — |
| Chess Puzzles | — | 14% |
| EnigmaEval | 2.7% | — |
| Mystery Game Puzzles | — | 8% |
| Surface Evolver Bench | — | 55% |
Math Too close to call
Gemini 2.5 Flash: 39.9 (#98), MiniMax-M3: 40.0 (#95)
| Benchmark | Gemini 2.5 Flash | MiniMax-M3 |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 73.1% | 71.1% |
| LMArena Math | 1415 | 1429 |
| ProofBench | — | 18% |
| Omni-MATH | 38.5% | — |
| FrontierMath (Feb 2025 set) | 4.8% | — |
| FrontierMath Tier 4 (v1) | 4.2% | — |
Knowledge MiniMax-M3 leads
Gemini 2.5 Flash: 36.4 (#168), MiniMax-M3: 58.4 (#35)
| Benchmark | Gemini 2.5 Flash | MiniMax-M3 |
|---|---|---|
| LMArena Expert | 1426 | 1461 |
| GPQA Diamond | — | 90.9% |
| Humanity's Last Exam | 12.1% | — |
| MMLU-Pro | 63.9% | — |
| Confabulations | 16.8% | — |
| Vectara Hallucination Rate | 7.8% | — |
| GPQA (HELM) | 39% | — |
Multimodal Gemini 2.5 Flash leads
Gemini 2.5 Flash: 41.8 (#32), MiniMax-M3: 40.2 (#51)
| Benchmark | Gemini 2.5 Flash | MiniMax-M3 |
|---|---|---|
| LMArena Vision | 1253 | 1253 |
| GeoBench | 76% | — |
| VPCT | 46.2% | — |
| LMArena Document | — | 1435 |
| SpatialViz-Bench | 36.9% | — |
Multilingual Too close to call
Gemini 2.5 Flash: 52.3 (#88), MiniMax-M3: 53.0 (#75)
| Benchmark | Gemini 2.5 Flash | MiniMax-M3 |
|---|---|---|
| LMArena Non-English | 1409 | 1420 |
| LMArena Chinese | 1450 | 1463 |
| LMArena French | 1433 | 1447 |
| LMArena German | 1418 | 1426 |
| LMArena Japanese | 1405 | 1381 |
| LMArena Korean | 1385 | 1372 |
| LMArena Russian | 1415 | 1428 |
| LMArena Spanish | 1421 | 1432 |
Instruction Following Too close to call
Gemini 2.5 Flash: 75.7 (#54), MiniMax-M3: 75.5 (#62)
| Benchmark | Gemini 2.5 Flash | MiniMax-M3 |
|---|---|---|
| LMArena Instruction Following | 1405 | 1433 |
| IFEval | 89.8% | — |
Long Context Gemini 2.5 Flash leads
Gemini 2.5 Flash: 47.5 (#17), MiniMax-M3: 44.2 (#72)
| Benchmark | Gemini 2.5 Flash | MiniMax-M3 |
|---|---|---|
| LMArena Longer Query | 1419 | 1445 |
| Fiction.LiveBench | 77.8% | — |
Writing & Preference MiniMax-M3 leads
Gemini 2.5 Flash: 53.8 (#157), MiniMax-M3: 62.1 (#83)
| Benchmark | Gemini 2.5 Flash | MiniMax-M3 |
|---|---|---|
| LMArena Text | 1417 | 1433 |
| LMArena Creative Writing | 1400 | 1404 |
| LMArena Multi-Turn | 1408 | 1442 |
| Short-Story Creative Writing | 76.5% | — |
| EQ-Bench Creative Writing | 1137 | — |
| WildBench | 81.7% | — |
| EQ-Bench 4 | — | 1150 |
Frequently asked questions
Is Gemini 2.5 Flash better than MiniMax-M3?
MiniMax-M3 is the stronger model overall, scoring 43.8 to 39.3 on the Noometry Index.
Which is cheaper, Gemini 2.5 Flash or MiniMax-M3?
MiniMax-M3 is cheaper. It lists at $0.30 per million input tokens and $1.20 per million output tokens; Gemini 2.5 Flash lists at $0.30 and $2.50.
Is Gemini 2.5 Flash or MiniMax-M3 better for coding?
MiniMax-M3 scores higher on coding benchmarks: 41.8 versus 35.8 in the Noometry coding category.
Which has the bigger context window?
Gemini 2.5 Flash does, with 1.05M tokens against 1M.
How many benchmarks do Gemini 2.5 Flash and MiniMax-M3 share?
27 benchmarks have published results for both models. Gemini 2.5 Flash has 54 scored results on Noometry and MiniMax-M3 has 41.