Model comparison
Llama 2-70B vs MiniMax-M3
MiniMax-M3 is the stronger model overall, scoring 43.8 to 24.4 on the Noometry Index.
Last verified . 22 shared benchmarks.
Meta
MiniMaxSummary
- They share 22 benchmarks with published results for both. Llama 2-70B scores higher in 0 categories and MiniMax-M3 in 8 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in knowledge, where MiniMax-M3 leads 58.4 to 7.4.
- The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 0% for Llama 2-70B and 71.1% for MiniMax-M3.
Side by side
| Llama 2-70B | MiniMax-M3 | |
|---|---|---|
| Provider | Meta | MiniMax |
| Noometry Index | 24.4 | 43.8 |
| Released | 2023-07-18 | 2026-06-01 |
| Weights | Open | Open |
| Context window | — | 1M |
| Max output | — | 512K |
| Input $ / M tokens | — | $0.30 |
| Output $ / M tokens | — | $1.20 |
| Results tracked | 35 | 41 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding MiniMax-M3 leads
Llama 2-70B: 31.4 (#286), MiniMax-M3: 41.8 (#118)
| Benchmark | Llama 2-70B | MiniMax-M3 |
|---|---|---|
| LMArena Coding | 1079 | 1469 |
| FrontierCode | — | 14.7% |
| LMArena WebDev | — | 1482 |
| SciCode | — | 47.1% |
| ALE-Bench | — | 640.02 |
Agentic & Tool Use Not comparable
Llama 2-70B: —, MiniMax-M3: 22.6 (#130)
| Benchmark | Llama 2-70B | MiniMax-M3 |
|---|---|---|
| APEX-Agents | — | 37.7% |
| OSWorld 2.0 | — | 4.6% |
| GBAEval | — | 0.9% |
| Vending-Bench 2 | — | 2,158 |
Reasoning MiniMax-M3 leads
Llama 2-70B: 14.4 (#325), MiniMax-M3: 30.1 (#87)
| Benchmark | Llama 2-70B | MiniMax-M3 |
|---|---|---|
| LMArena Hard Prompts | 1073 | 1447 |
| DTBench | 41.6% | 78.9% |
| Epoch Capabilities Index | 113.79 | 146.95 |
| ForecastBench | 51.4 | 61.4 |
| SimpleBench | — | 45.8% |
| NYT Connections (extended) | — | 65.1% |
| CritPt | — | 3.7% |
| Chess Puzzles | — | 14% |
| Mystery Game Puzzles | — | 8% |
| LMCA | — | 33.7% |
| Surface Evolver Bench | — | 55% |
| BIG-Bench Hard | 64.9% | — |
| CommonsenseQA 2.0 | 50% | — |
| HellaSwag | 85.3% | — |
| LAMBADA | 78.9% | — |
| PIQA | 82.8% | — |
| WinoGrande | 80.2% | — |
Math MiniMax-M3 leads
Llama 2-70B: 8.1 (#326), MiniMax-M3: 40.0 (#95)
| Benchmark | Llama 2-70B | MiniMax-M3 |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 0% | 71.1% |
| LMArena Math | 1091 | 1429 |
| ProofBench | — | 18% |
| MATH Level 5 | 3.3% | — |
| GSM8K | 69.6% | — |
Knowledge MiniMax-M3 leads
Llama 2-70B: 7.4 (#310), MiniMax-M3: 58.4 (#35)
| Benchmark | Llama 2-70B | MiniMax-M3 |
|---|---|---|
| GPQA Diamond | 26.3% | 90.9% |
| LMArena Expert | 1039 | 1461 |
| ARC (AI2) Challenge | 78.3% | — |
| BoolQ | 88.6% | — |
| MMLU | 69.9% | — |
| OpenBookQA | 60.2% | — |
| TriviaQA | 87.6% | — |
Multimodal Not comparable
Llama 2-70B: —, MiniMax-M3: 40.2 (#51)
| Benchmark | Llama 2-70B | MiniMax-M3 |
|---|---|---|
| LMArena Vision | — | 1253 |
| LMArena Document | — | 1435 |
Multilingual MiniMax-M3 leads
Llama 2-70B: 27.7 (#274), MiniMax-M3: 53.0 (#75)
| Benchmark | Llama 2-70B | MiniMax-M3 |
|---|---|---|
| LMArena Non-English | 1045 | 1420 |
| LMArena Chinese | 995 | 1463 |
| LMArena French | 1090 | 1447 |
| LMArena German | 1041 | 1426 |
| LMArena Japanese | 927 | 1381 |
| LMArena Korean | 964 | 1372 |
| LMArena Russian | 1083 | 1428 |
| LMArena Spanish | 1143 | 1432 |
Instruction Following MiniMax-M3 leads
Llama 2-70B: 54.9 (#278), MiniMax-M3: 75.5 (#62)
| Benchmark | Llama 2-70B | MiniMax-M3 |
|---|---|---|
| LMArena Instruction Following | 1071 | 1433 |
Long Context MiniMax-M3 leads
Llama 2-70B: 32.3 (#270), MiniMax-M3: 44.2 (#72)
| Benchmark | Llama 2-70B | MiniMax-M3 |
|---|---|---|
| LMArena Longer Query | 1062 | 1445 |
Writing & Preference MiniMax-M3 leads
Llama 2-70B: 32.3 (#279), MiniMax-M3: 62.1 (#83)
| Benchmark | Llama 2-70B | MiniMax-M3 |
|---|---|---|
| LMArena Text | 1115 | 1433 |
| LMArena Creative Writing | 1075 | 1404 |
| LMArena Multi-Turn | 1088 | 1442 |
| EQ-Bench 4 | — | 1150 |
Frequently asked questions
Is Llama 2-70B better than MiniMax-M3?
MiniMax-M3 is the stronger model overall, scoring 43.8 to 24.4 on the Noometry Index.
Is Llama 2-70B or MiniMax-M3 better for coding?
MiniMax-M3 scores higher on coding benchmarks: 41.8 versus 31.4 in the Noometry coding category.
How many benchmarks do Llama 2-70B and MiniMax-M3 share?
22 benchmarks have published results for both models. Llama 2-70B has 35 scored results on Noometry and MiniMax-M3 has 41.