Model comparison
Llama 4 Maverick vs MiniMax-M2.5
MiniMax-M2.5 is the stronger model overall, scoring 38.1 to 30.6 on the Noometry Index. Llama 4 Maverick costs 1.7× less per token, which makes it the better buy when MiniMax-M2.5's lead doesn't matter for your workload.
Last verified . 26 shared benchmarks.
Meta
MiniMaxSummary
- They share 26 benchmarks with published results for both. Llama 4 Maverick scores higher in 1 category and MiniMax-M2.5 in 8 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in coding, where MiniMax-M2.5 leads 47.9 to 26.3.
- The biggest single-benchmark swing is ARC-AGI-1: 4.4% for Llama 4 Maverick and 63.7% for MiniMax-M2.5.
- Llama 4 Maverick is cheaper at $0.19 / $0.65 per million input/output tokens, against $0.30 / $1.20 for MiniMax-M2.5.
- MiniMax-M2.5 accepts more context: 205K tokens versus 128K.
Side by side
| Llama 4 Maverick | MiniMax-M2.5 | |
|---|---|---|
| Provider | Meta | MiniMax |
| Noometry Index | 30.6 | 38.1 |
| Released | 2025-04-05 | 2026-02-12 |
| Weights | Open | Open |
| Context window | 128K | 205K |
| Max output | 4K | 131K |
| Input $ / M tokens | $0.19 | $0.30 |
| Output $ / M tokens | $0.65 | $1.20 |
| Results tracked | 54 | 33 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding MiniMax-M2.5 leads
Llama 4 Maverick: 26.3 (#318), MiniMax-M2.5: 47.9 (#53)
| Benchmark | Llama 4 Maverick | MiniMax-M2.5 |
|---|---|---|
| SWE-bench Verified (bash only) | 21% | 75.8% |
| LMArena Coding | 1302 | 1380 |
| ALE-Bench | 172.97 | 618.17 |
| Aider Polyglot | 15.6% | — |
| LMArena WebDev | — | 1386 |
| SWE-bench Multilingual | — | 68.3% |
| SciCode | 33.1% | — |
| WeirdML | 24.5% | — |
| BigCodeBench Instruct | 49.7% | — |
| BigCodeBench Complete | 61.4% | — |
Agentic & Tool Use MiniMax-M2.5 leads
Llama 4 Maverick: 27.9 (#86), MiniMax-M2.5: 30.2 (#70)
| Benchmark | Llama 4 Maverick | MiniMax-M2.5 |
|---|---|---|
| Terminal-Bench | — | 42.7% |
| Berkeley Function Calling Leaderboard | 37.3% | — |
| Vending-Bench 2 | — | -23.16 |
Reasoning MiniMax-M2.5 leads
Llama 4 Maverick: 10.1 (#337), MiniMax-M2.5: 17.6 (#281)
| Benchmark | Llama 4 Maverick | MiniMax-M2.5 |
|---|---|---|
| ARC-AGI-2 | 0% | 4.9% |
| Kagi LLM Benchmark | 55.9% | 55.2% |
| NYT Connections (extended) | 8% | 16.8% |
| ARC-AGI-1 | 4.4% | 63.7% |
| LMArena Hard Prompts | 1281 | 1372 |
| Epoch Capabilities Index | 132.2 | 146.68 |
| SimpleBench | 27.7% | — |
| CritPt | 0% | — |
| EnigmaEval | 0.6% | — |
| DTBench | 61.9% | — |
| LMCA | 15.9% | — |
| ForecastBench | 57.5 | — |
Math MiniMax-M2.5 leads
Llama 4 Maverick: 24.6 (#256), MiniMax-M2.5: 25.8 (#242)
| Benchmark | Llama 4 Maverick | MiniMax-M2.5 |
|---|---|---|
| LMArena Math | 1299 | 1378 |
| OTIS Mock AIME 2024-2025 | 20.6% | — |
| ProofBench | — | 4% |
| Omni-MATH | 42.2% | — |
| MATH Level 5 | 73% | — |
| FrontierMath (Feb 2025 set) | 0.7% | — |
Knowledge MiniMax-M2.5 leads
Llama 4 Maverick: 33.3 (#196), MiniMax-M2.5: 38.6 (#132)
| Benchmark | Llama 4 Maverick | MiniMax-M2.5 |
|---|---|---|
| Vectara Hallucination Rate | 8.2% | 9.1% |
| LMArena Expert | 1259 | 1378 |
| GPQA Diamond | 67% | — |
| Humanity's Last Exam | 5.7% | — |
| MMLU-Pro | 81% | — |
| Confabulations | 22.6% | — |
| GPQA (HELM) | 65% | — |
Multimodal Not comparable
Llama 4 Maverick: 30.9 (#100), MiniMax-M2.5: —
| Benchmark | Llama 4 Maverick | MiniMax-M2.5 |
|---|---|---|
| LMArena Vision | 1141 | — |
| GeoBench | 52% | — |
| SpatialViz-Bench | 31.8% | — |
Multilingual MiniMax-M2.5 leads
Llama 4 Maverick: 42.4 (#190), MiniMax-M2.5: 47.3 (#147)
| Benchmark | Llama 4 Maverick | MiniMax-M2.5 |
|---|---|---|
| LMArena Non-English | 1269 | 1338 |
| LMArena Chinese | 1277 | 1392 |
| LMArena French | 1258 | 1363 |
| LMArena German | 1291 | 1361 |
| LMArena Japanese | 1207 | 1169 |
| LMArena Korean | 1203 | 1232 |
| LMArena Russian | 1286 | 1359 |
| LMArena Spanish | 1293 | 1354 |
Instruction Following Too close to call
Llama 4 Maverick: 71.9 (#140), MiniMax-M2.5: 71.7 (#143)
| Benchmark | Llama 4 Maverick | MiniMax-M2.5 |
|---|---|---|
| LMArena Instruction Following | 1267 | 1353 |
| IFEval | 90.8% | — |
Long Context MiniMax-M2.5 leads
Llama 4 Maverick: 31.4 (#274), MiniMax-M2.5: 37.4 (#211)
| Benchmark | Llama 4 Maverick | MiniMax-M2.5 |
|---|---|---|
| LMArena Longer Query | 1280 | 1367 |
| Fiction.LiveBench | 46.2% | — |
| CL-bench | — | 11.4% |
| CL-bench Life | — | 6.3% |
Writing & Preference MiniMax-M2.5 leads
Llama 4 Maverick: 39.0 (#247), MiniMax-M2.5: 54.1 (#149)
| Benchmark | Llama 4 Maverick | MiniMax-M2.5 |
|---|---|---|
| LMArena Text | 1287 | 1359 |
| LMArena Creative Writing | 1266 | 1331 |
| EQ-Bench Creative Writing | 860 | 1361 |
| LMArena Multi-Turn | 1289 | 1364 |
| Short-Story Creative Writing | 62% | — |
| WildBench | 80% | — |
Frequently asked questions
Is Llama 4 Maverick better than MiniMax-M2.5?
MiniMax-M2.5 is the stronger model overall, scoring 38.1 to 30.6 on the Noometry Index. Llama 4 Maverick costs 1.7× less per token, which makes it the better buy when MiniMax-M2.5's lead doesn't matter for your workload.
Which is cheaper, Llama 4 Maverick or MiniMax-M2.5?
Llama 4 Maverick is cheaper. It lists at $0.19 per million input tokens and $0.65 per million output tokens; MiniMax-M2.5 lists at $0.30 and $1.20.
Is Llama 4 Maverick or MiniMax-M2.5 better for coding?
MiniMax-M2.5 scores higher on coding benchmarks: 47.9 versus 26.3 in the Noometry coding category.
Which has the bigger context window?
MiniMax-M2.5 does, with 205K tokens against 128K.
How many benchmarks do Llama 4 Maverick and MiniMax-M2.5 share?
26 benchmarks have published results for both models. Llama 4 Maverick has 54 scored results on Noometry and MiniMax-M2.5 has 33.