Model comparison
Gemma 4 31B IT vs MiniMax-M2.5
Gemma 4 31B IT is the stronger model overall, scoring 43.2 to 38.1 on the Noometry Index.
Last verified . 21 shared benchmarks.
Google
MiniMaxSummary
- They share 21 benchmarks with published results for both. Gemma 4 31B IT scores higher in 6 categories and MiniMax-M2.5 in 2 categories; 7 gaps are clear of the uncertainty.
- The widest gap is in math, where Gemma 4 31B IT leads 41.3 to 25.8.
- The biggest single-benchmark swing is NYT Connections (extended): 70.6% for Gemma 4 31B IT and 16.8% for MiniMax-M2.5.
- Gemma 4 31B IT is cheaper at $0.09 / $0.34 per million input/output tokens, against $0.30 / $1.20 for MiniMax-M2.5.
- Gemma 4 31B IT accepts more context: 262K tokens versus 205K.
Side by side
| Gemma 4 31B IT | MiniMax-M2.5 | |
|---|---|---|
| Provider | MiniMax | |
| Noometry Index | 43.2 | 38.1 |
| Released | 2026-04-02 | 2026-02-12 |
| Weights | Open | Open |
| Context window | 262K | 205K |
| Max output | 33K | 131K |
| Input $ / M tokens | $0.09 | $0.30 |
| Output $ / M tokens | $0.34 | $1.20 |
| Results tracked | 35 | 33 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding MiniMax-M2.5 leads
Gemma 4 31B IT: 41.9 (#98), MiniMax-M2.5: 47.9 (#53)
| Benchmark | Gemma 4 31B IT | MiniMax-M2.5 |
|---|---|---|
| LMArena WebDev | 1364 | 1386 |
| LMArena Coding | 1459 | 1380 |
| ALE-Bench | 925.5 | 618.17 |
| SWE-bench Verified (bash only) | — | 75.8% |
| SWE-bench Multilingual | — | 68.3% |
| SciCode | 43.4% | — |
| WeirdML | 52.3% | — |
Agentic & Tool Use Not comparable
Gemma 4 31B IT: —, MiniMax-M2.5: 30.2 (#70)
| Benchmark | Gemma 4 31B IT | MiniMax-M2.5 |
|---|---|---|
| Terminal-Bench | — | 42.7% |
| Vending-Bench 2 | — | -23.16 |
Reasoning Gemma 4 31B IT leads
Gemma 4 31B IT: 26.4 (#119), MiniMax-M2.5: 17.6 (#281)
| Benchmark | Gemma 4 31B IT | MiniMax-M2.5 |
|---|---|---|
| Kagi LLM Benchmark | 63.5% | 55.2% |
| NYT Connections (extended) | 70.6% | 16.8% |
| LMArena Hard Prompts | 1448 | 1372 |
| Epoch Capabilities Index | 142.74 | 146.68 |
| ARC-AGI-2 | — | 4.9% |
| ARC-AGI-1 | — | 63.7% |
| CritPt | 1.4% | — |
| Chess Puzzles | 5% | — |
| Thematic Generalization | 53% | — |
| DTBench | 82.7% | — |
| LMCA | 39.3% | — |
| Surface Evolver Bench | 30.6% | — |
Math Gemma 4 31B IT leads
Gemma 4 31B IT: 41.3 (#72), MiniMax-M2.5: 25.8 (#242)
| Benchmark | Gemma 4 31B IT | MiniMax-M2.5 |
|---|---|---|
| LMArena Math | 1464 | 1378 |
| OTIS Mock AIME 2024-2025 | 73.3% | — |
| ProofBench | — | 4% |
Knowledge Too close to call
Gemma 4 31B IT: 37.9 (#141), MiniMax-M2.5: 38.6 (#132)
| Benchmark | Gemma 4 31B IT | MiniMax-M2.5 |
|---|---|---|
| Vectara Hallucination Rate | 7.4% | 9.1% |
| LMArena Expert | 1465 | 1378 |
| GPQA Diamond | 75.8% | — |
| SimpleQA Verified | 10.4% | — |
Multimodal Not comparable
Gemma 4 31B IT: 41.0 (#32), MiniMax-M2.5: —
| Benchmark | Gemma 4 31B IT | MiniMax-M2.5 |
|---|---|---|
| LMArena Vision | 1278 | — |
| LMArena Document | 1425 | — |
Multilingual Gemma 4 31B IT leads
Gemma 4 31B IT: 54.0 (#52), MiniMax-M2.5: 47.3 (#147)
| Benchmark | Gemma 4 31B IT | MiniMax-M2.5 |
|---|---|---|
| LMArena Non-English | 1431 | 1338 |
| LMArena Chinese | 1478 | 1392 |
| LMArena French | 1436 | 1363 |
| LMArena Russian | 1462 | 1359 |
| LMArena Spanish | 1443 | 1354 |
| LMArena German | — | 1361 |
| LMArena Japanese | — | 1169 |
| LMArena Korean | — | 1232 |
Instruction Following Gemma 4 31B IT leads
Gemma 4 31B IT: 75.6 (#58), MiniMax-M2.5: 71.7 (#143)
| Benchmark | Gemma 4 31B IT | MiniMax-M2.5 |
|---|---|---|
| LMArena Instruction Following | 1433 | 1353 |
Long Context Gemma 4 31B IT leads
Gemma 4 31B IT: 44.2 (#67), MiniMax-M2.5: 37.4 (#211)
| Benchmark | Gemma 4 31B IT | MiniMax-M2.5 |
|---|---|---|
| LMArena Longer Query | 1446 | 1367 |
| CL-bench | — | 11.4% |
| CL-bench Life | — | 6.3% |
Writing & Preference Gemma 4 31B IT leads
Gemma 4 31B IT: 60.7 (#92), MiniMax-M2.5: 54.1 (#149)
| Benchmark | Gemma 4 31B IT | MiniMax-M2.5 |
|---|---|---|
| LMArena Text | 1443 | 1359 |
| LMArena Creative Writing | 1416 | 1331 |
| EQ-Bench Creative Writing | 1368 | 1361 |
| LMArena Multi-Turn | 1452 | 1364 |
| EQ-Bench 4 | 1120 | — |
Frequently asked questions
Is Gemma 4 31B IT better than MiniMax-M2.5?
Gemma 4 31B IT is the stronger model overall, scoring 43.2 to 38.1 on the Noometry Index.
Which is cheaper, Gemma 4 31B IT or MiniMax-M2.5?
Gemma 4 31B IT is cheaper. It lists at $0.09 per million input tokens and $0.34 per million output tokens; MiniMax-M2.5 lists at $0.30 and $1.20.
Is Gemma 4 31B IT or MiniMax-M2.5 better for coding?
MiniMax-M2.5 scores higher on coding benchmarks: 47.9 versus 41.9 in the Noometry coding category.
Which has the bigger context window?
Gemma 4 31B IT does, with 262K tokens against 205K.
How many benchmarks do Gemma 4 31B IT and MiniMax-M2.5 share?
21 benchmarks have published results for both models. Gemma 4 31B IT has 35 scored results on Noometry and MiniMax-M2.5 has 33.