Model comparison
Llama 4 Scout vs MiniMax-M3
MiniMax-M3 is the stronger model overall, scoring 43.8 to 27.7 on the Noometry Index. Llama 4 Scout costs 3.5× less per token, which makes it the better buy when MiniMax-M3'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 Scout scores higher in 1 category and MiniMax-M3 in 9 categories; 10 gaps are clear of the uncertainty.
- The widest gap is in knowledge, where MiniMax-M3 leads 58.4 to 31.9.
- The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 7.8% for Llama 4 Scout and 71.1% for MiniMax-M3.
- Llama 4 Scout is cheaper at $0.10 / $0.30 per million input/output tokens, against $0.30 / $1.20 for MiniMax-M3.
- MiniMax-M3 accepts more context: 1M tokens versus 128K.
Side by side
| Llama 4 Scout | MiniMax-M3 | |
|---|---|---|
| Provider | Meta | MiniMax |
| Noometry Index | 27.7 | 43.8 |
| Released | 2025-04-05 | 2026-06-01 |
| Weights | Open | Open |
| Context window | 128K | 1M |
| Max output | 4K | 512K |
| Input $ / M tokens | $0.10 | $0.30 |
| Output $ / M tokens | $0.30 | $1.20 |
| Results tracked | 43 | 41 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding MiniMax-M3 leads
Llama 4 Scout: 20.2 (#339), MiniMax-M3: 41.8 (#118)
| Benchmark | Llama 4 Scout | MiniMax-M3 |
|---|---|---|
| SciCode | 17% | 47.1% |
| LMArena Coding | 1286 | 1469 |
| FrontierCode | — | 14.7% |
| SWE-bench Verified (bash only) | 9.1% | — |
| LMArena WebDev | — | 1482 |
| BigCodeBench Complete | 43.1% | — |
| ALE-Bench | — | 640.02 |
Agentic & Tool Use Llama 4 Scout leads
Llama 4 Scout: 24.6 (#119), MiniMax-M3: 22.6 (#130)
| Benchmark | Llama 4 Scout | MiniMax-M3 |
|---|---|---|
| APEX-Agents | — | 37.7% |
| Berkeley Function Calling Leaderboard | 28.1% | — |
| OSWorld 2.0 | — | 4.6% |
| GBAEval | — | 0.9% |
| Vending-Bench 2 | — | 2,158 |
Reasoning MiniMax-M3 leads
Llama 4 Scout: 9.1 (#345), MiniMax-M3: 30.1 (#87)
| Benchmark | Llama 4 Scout | MiniMax-M3 |
|---|---|---|
| CritPt | 0% | 3.7% |
| LMArena Hard Prompts | 1266 | 1447 |
| DTBench | 57.9% | 78.9% |
| LMCA | 12% | 33.7% |
| Epoch Capabilities Index | 129.64 | 146.95 |
| ForecastBench | 57.5 | 61.4 |
| ARC-AGI-2 | 0% | — |
| SimpleBench | — | 45.8% |
| Kagi LLM Benchmark | 36.9% | — |
| NYT Connections (extended) | — | 65.1% |
| ARC-AGI-1 | 0.5% | — |
| Chess Puzzles | — | 14% |
| Mystery Game Puzzles | — | 8% |
| Surface Evolver Bench | — | 55% |
Math MiniMax-M3 leads
Llama 4 Scout: 19.6 (#286), MiniMax-M3: 40.0 (#95)
| Benchmark | Llama 4 Scout | MiniMax-M3 |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 7.8% | 71.1% |
| LMArena Math | 1287 | 1429 |
| ProofBench | — | 18% |
| Omni-MATH | 37.3% | — |
| MATH Level 5 | 62.3% | — |
| FrontierMath (Feb 2025 set) | 0% | — |
Knowledge MiniMax-M3 leads
Llama 4 Scout: 31.9 (#217), MiniMax-M3: 58.4 (#35)
| Benchmark | Llama 4 Scout | MiniMax-M3 |
|---|---|---|
| GPQA Diamond | 51.8% | 90.9% |
| LMArena Expert | 1235 | 1461 |
| MMLU-Pro | 74.2% | — |
| Vectara Hallucination Rate | 7.7% | — |
| GPQA (HELM) | 50.7% | — |
Multimodal MiniMax-M3 leads
Llama 4 Scout: 32.2 (#102), MiniMax-M3: 40.2 (#51)
| Benchmark | Llama 4 Scout | MiniMax-M3 |
|---|---|---|
| LMArena Vision | 1118 | 1253 |
| LMArena Document | — | 1435 |
| SpatialViz-Bench | 34.2% | — |
Multilingual MiniMax-M3 leads
Llama 4 Scout: 41.0 (#212), MiniMax-M3: 53.0 (#75)
| Benchmark | Llama 4 Scout | MiniMax-M3 |
|---|---|---|
| LMArena Non-English | 1252 | 1420 |
| LMArena Chinese | 1255 | 1463 |
| LMArena French | 1282 | 1447 |
| LMArena German | 1272 | 1426 |
| LMArena Japanese | 1206 | 1381 |
| LMArena Korean | 1207 | 1372 |
| LMArena Russian | 1263 | 1428 |
| LMArena Spanish | 1278 | 1432 |
Instruction Following MiniMax-M3 leads
Llama 4 Scout: 65.8 (#217), MiniMax-M3: 75.5 (#62)
| Benchmark | Llama 4 Scout | MiniMax-M3 |
|---|---|---|
| LMArena Instruction Following | 1248 | 1433 |
| IFEval | 81.8% | — |
Long Context MiniMax-M3 leads
Llama 4 Scout: 27.5 (#294), MiniMax-M3: 44.2 (#72)
| Benchmark | Llama 4 Scout | MiniMax-M3 |
|---|---|---|
| LMArena Longer Query | 1265 | 1445 |
| Fiction.LiveBench | 36% | — |
Writing & Preference MiniMax-M3 leads
Llama 4 Scout: 37.0 (#261), MiniMax-M3: 62.1 (#83)
| Benchmark | Llama 4 Scout | MiniMax-M3 |
|---|---|---|
| LMArena Text | 1279 | 1433 |
| LMArena Creative Writing | 1249 | 1404 |
| LMArena Multi-Turn | 1280 | 1442 |
| EQ-Bench Creative Writing | 783 | — |
| WildBench | 78% | — |
| EQ-Bench 4 | — | 1150 |
Frequently asked questions
Is Llama 4 Scout better than MiniMax-M3?
MiniMax-M3 is the stronger model overall, scoring 43.8 to 27.7 on the Noometry Index. Llama 4 Scout costs 3.5× less per token, which makes it the better buy when MiniMax-M3's lead doesn't matter for your workload.
Which is cheaper, Llama 4 Scout or MiniMax-M3?
Llama 4 Scout is cheaper. It lists at $0.10 per million input tokens and $0.30 per million output tokens; MiniMax-M3 lists at $0.30 and $1.20.
Is Llama 4 Scout or MiniMax-M3 better for coding?
MiniMax-M3 scores higher on coding benchmarks: 41.8 versus 20.2 in the Noometry coding category.
Which has the bigger context window?
MiniMax-M3 does, with 1M tokens against 128K.
How many benchmarks do Llama 4 Scout and MiniMax-M3 share?
26 benchmarks have published results for both models. Llama 4 Scout has 43 scored results on Noometry and MiniMax-M3 has 41.