Model comparison
Llama 4 Scout vs MiniMax-M2.5
MiniMax-M2.5 is the stronger model overall, scoring 38.1 to 27.4 on the Noometry Index. Llama 4 Scout costs 3.5× less per token, which makes it the better buy when MiniMax-M2.5's lead doesn't matter for your workload.
Last verified . 24 shared benchmarks.
Meta
MiniMaxSummary
- They share 24 benchmarks with published results for both. Llama 4 Scout scores higher in 0 categories and MiniMax-M2.5 in 9 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in coding, where MiniMax-M2.5 leads 47.9 to 19.9.
- The biggest single-benchmark swing is SWE-bench Verified (bash only): 9.1% for Llama 4 Scout and 75.8% for MiniMax-M2.5.
- Llama 4 Scout is cheaper at $0.10 / $0.30 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 Scout | MiniMax-M2.5 | |
|---|---|---|
| Provider | Meta | MiniMax |
| Noometry Index | 27.4 | 38.1 |
| Released | 2025-04-05 | 2026-02-12 |
| Weights | Open | Open |
| Context window | 128K | 205K |
| Max output | 4K | 131K |
| Input $ / M tokens | $0.10 | $0.30 |
| Output $ / M tokens | $0.30 | $1.20 |
| Results tracked | 43 | 33 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding MiniMax-M2.5 leads
Llama 4 Scout: 19.9 (#334), MiniMax-M2.5: 47.9 (#53)
| Benchmark | Llama 4 Scout | MiniMax-M2.5 |
|---|---|---|
| SWE-bench Verified (bash only) | 9.1% | 75.8% |
| LMArena Coding | 1286 | 1380 |
| LMArena WebDev | — | 1386 |
| SWE-bench Multilingual | — | 68.3% |
| SciCode | 17% | — |
| BigCodeBench Complete | 43.1% | — |
| ALE-Bench | — | 618.17 |
Agentic & Tool Use MiniMax-M2.5 leads
Llama 4 Scout: 24.4 (#109), MiniMax-M2.5: 30.2 (#70)
| Benchmark | Llama 4 Scout | MiniMax-M2.5 |
|---|---|---|
| Terminal-Bench | — | 42.7% |
| Berkeley Function Calling Leaderboard | 28.1% | — |
| Vending-Bench 2 | — | -23.16 |
Reasoning MiniMax-M2.5 leads
Llama 4 Scout: 9.2 (#339), MiniMax-M2.5: 17.6 (#281)
| Benchmark | Llama 4 Scout | MiniMax-M2.5 |
|---|---|---|
| ARC-AGI-2 | 0% | 4.9% |
| Kagi LLM Benchmark | 36.9% | 55.2% |
| ARC-AGI-1 | 0.5% | 63.7% |
| LMArena Hard Prompts | 1266 | 1372 |
| Epoch Capabilities Index | 129.64 | 146.68 |
| NYT Connections (extended) | — | 16.8% |
| CritPt | 0% | — |
| DTBench | 57.9% | — |
| LMCA | 12% | — |
| ForecastBench | 57.5 | — |
Math MiniMax-M2.5 leads
Llama 4 Scout: 18.4 (#280), MiniMax-M2.5: 25.8 (#242)
| Benchmark | Llama 4 Scout | MiniMax-M2.5 |
|---|---|---|
| LMArena Math | 1286 | 1378 |
| OTIS Mock AIME 2024-2025 | 7.8% | — |
| ProofBench | — | 4% |
| Omni-MATH | 37.3% | — |
| MATH Level 5 | 62.3% | — |
| FrontierMath (Feb 2025 set) | 0% | — |
Knowledge MiniMax-M2.5 leads
Llama 4 Scout: 31.8 (#208), MiniMax-M2.5: 38.6 (#132)
| Benchmark | Llama 4 Scout | MiniMax-M2.5 |
|---|---|---|
| Vectara Hallucination Rate | 7.7% | 9.1% |
| LMArena Expert | 1236 | 1378 |
| GPQA Diamond | 51.8% | — |
| MMLU-Pro | 74.2% | — |
| GPQA (HELM) | 50.7% | — |
Multimodal Not comparable
Llama 4 Scout: 31.6 (#98), MiniMax-M2.5: —
| Benchmark | Llama 4 Scout | MiniMax-M2.5 |
|---|---|---|
| LMArena Vision | 1118 | — |
| SpatialViz-Bench | 34.2% | — |
Multilingual MiniMax-M2.5 leads
Llama 4 Scout: 41.3 (#207), MiniMax-M2.5: 47.3 (#147)
| Benchmark | Llama 4 Scout | MiniMax-M2.5 |
|---|---|---|
| LMArena Non-English | 1253 | 1338 |
| LMArena Chinese | 1256 | 1392 |
| LMArena French | 1283 | 1363 |
| LMArena German | 1274 | 1361 |
| LMArena Japanese | 1209 | 1169 |
| LMArena Korean | 1207 | 1232 |
| LMArena Russian | 1263 | 1359 |
| LMArena Spanish | 1277 | 1354 |
Instruction Following MiniMax-M2.5 leads
Llama 4 Scout: 66.0 (#212), MiniMax-M2.5: 71.7 (#143)
| Benchmark | Llama 4 Scout | MiniMax-M2.5 |
|---|---|---|
| LMArena Instruction Following | 1248 | 1353 |
| IFEval | 81.8% | — |
Long Context MiniMax-M2.5 leads
Llama 4 Scout: 27.5 (#289), MiniMax-M2.5: 37.4 (#211)
| Benchmark | Llama 4 Scout | MiniMax-M2.5 |
|---|---|---|
| LMArena Longer Query | 1265 | 1367 |
| Fiction.LiveBench | 36% | — |
| CL-bench | — | 11.4% |
| CL-bench Life | — | 6.3% |
Writing & Preference MiniMax-M2.5 leads
Llama 4 Scout: 37.1 (#256), MiniMax-M2.5: 54.1 (#149)
| Benchmark | Llama 4 Scout | MiniMax-M2.5 |
|---|---|---|
| LMArena Text | 1279 | 1359 |
| LMArena Creative Writing | 1249 | 1331 |
| EQ-Bench Creative Writing | 783 | 1361 |
| LMArena Multi-Turn | 1281 | 1364 |
| WildBench | 78% | — |
Frequently asked questions
Is Llama 4 Scout better than MiniMax-M2.5?
MiniMax-M2.5 is the stronger model overall, scoring 38.1 to 27.4 on the Noometry Index. Llama 4 Scout costs 3.5× 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 Scout or MiniMax-M2.5?
Llama 4 Scout is cheaper. It lists at $0.10 per million input tokens and $0.30 per million output tokens; MiniMax-M2.5 lists at $0.30 and $1.20.
Is Llama 4 Scout or MiniMax-M2.5 better for coding?
MiniMax-M2.5 scores higher on coding benchmarks: 47.9 versus 19.9 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 Scout and MiniMax-M2.5 share?
24 benchmarks have published results for both models. Llama 4 Scout has 43 scored results on Noometry and MiniMax-M2.5 has 33.