Model comparison
Llama 2-70B vs MiniMax-M2.7
MiniMax-M2.7 is the stronger model overall, scoring 37.4 to 24.1 on the Noometry Index.
Last verified . 18 shared benchmarks.
Meta
MiniMaxSummary
- They share 18 benchmarks with published results for both. Llama 2-70B scores higher in 0 categories and MiniMax-M2.7 in 8 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in knowledge, where MiniMax-M2.7 leads 37.1 to 7.4.
Side by side
| Llama 2-70B | MiniMax-M2.7 | |
|---|---|---|
| Provider | Meta | MiniMax |
| Noometry Index | 24.1 | 37.4 |
| Released | 2023-07-18 | 2026-03-18 |
| Weights | Open | Open |
| Context window | — | 205K |
| Max output | — | 131K |
| Input $ / M tokens | — | $0.30 |
| Output $ / M tokens | — | $1.20 |
| Results tracked | 35 | 30 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding MiniMax-M2.7 leads
Llama 2-70B: 30.8 (#281), MiniMax-M2.7: 41.4 (#115)
| Benchmark | Llama 2-70B | MiniMax-M2.7 |
|---|---|---|
| LMArena Coding | 1079 | 1454 |
| LMArena WebDev | — | 1397 |
| SciCode | — | 47% |
| WeirdML | — | 37% |
| ALE-Bench | — | 599.25 |
Agentic & Tool Use Not comparable
Llama 2-70B: —, MiniMax-M2.7: 24.9 (#103)
| Benchmark | Llama 2-70B | MiniMax-M2.7 |
|---|---|---|
| Terminal-Bench | — | 45.1% |
| ExploitBench | — | 13.3% |
| GBAEval | — | 0% |
Reasoning MiniMax-M2.7 leads
Llama 2-70B: 13.7 (#324), MiniMax-M2.7: 19.0 (#248)
| Benchmark | Llama 2-70B | MiniMax-M2.7 |
|---|---|---|
| LMArena Hard Prompts | 1073 | 1422 |
| Epoch Capabilities Index | 113.59 | 145.84 |
| NYT Connections (extended) | — | 24.7% |
| CritPt | — | 0.6% |
| Thematic Generalization | — | 39.3% |
| DTBench | 41.6% | — |
| BIG-Bench Hard | 64.9% | — |
| CommonsenseQA 2.0 | 50% | — |
| ForecastBench | 51.4 | — |
| HellaSwag | 85.3% | — |
| LAMBADA | 78.9% | — |
| PIQA | 82.8% | — |
| WinoGrande | 80.2% | — |
Math MiniMax-M2.7 leads
Llama 2-70B: 7.5 (#320), MiniMax-M2.7: 24.9 (#254)
| Benchmark | Llama 2-70B | MiniMax-M2.7 |
|---|---|---|
| LMArena Math | 1091 | 1420 |
| OTIS Mock AIME 2024-2025 | 0% | — |
| ProofBench | — | 3% |
| MATH Level 5 | 3.3% | — |
| GSM8K | 69.6% | — |
Knowledge MiniMax-M2.7 leads
Llama 2-70B: 7.4 (#305), MiniMax-M2.7: 37.1 (#151)
| Benchmark | Llama 2-70B | MiniMax-M2.7 |
|---|---|---|
| LMArena Expert | 1039 | 1443 |
| GPQA Diamond | 26.3% | — |
| Vectara Hallucination Rate | — | 12.9% |
| ARC (AI2) Challenge | 78.3% | — |
| BoolQ | 88.6% | — |
| MMLU | 69.9% | — |
| OpenBookQA | 60.2% | — |
| TriviaQA | 87.6% | — |
Multilingual MiniMax-M2.7 leads
Llama 2-70B: 27.9 (#269), MiniMax-M2.7: 50.5 (#120)
| Benchmark | Llama 2-70B | MiniMax-M2.7 |
|---|---|---|
| LMArena Non-English | 1045 | 1382 |
| LMArena Chinese | 994 | 1439 |
| LMArena French | 1089 | 1422 |
| LMArena German | 1041 | 1398 |
| LMArena Japanese | 927 | 1261 |
| LMArena Korean | 964 | 1313 |
| LMArena Russian | 1083 | 1383 |
| LMArena Spanish | 1142 | 1403 |
Instruction Following MiniMax-M2.7 leads
Llama 2-70B: 55.1 (#273), MiniMax-M2.7: 74.2 (#101)
| Benchmark | Llama 2-70B | MiniMax-M2.7 |
|---|---|---|
| LMArena Instruction Following | 1071 | 1404 |
Long Context MiniMax-M2.7 leads
Llama 2-70B: 32.2 (#265), MiniMax-M2.7: 43.3 (#94)
| Benchmark | Llama 2-70B | MiniMax-M2.7 |
|---|---|---|
| LMArena Longer Query | 1062 | 1419 |
Writing & Preference MiniMax-M2.7 leads
Llama 2-70B: 32.5 (#274), MiniMax-M2.7: 59.1 (#107)
| Benchmark | Llama 2-70B | MiniMax-M2.7 |
|---|---|---|
| LMArena Text | 1115 | 1404 |
| LMArena Creative Writing | 1075 | 1354 |
| LMArena Multi-Turn | 1088 | 1413 |
Frequently asked questions
Is Llama 2-70B better than MiniMax-M2.7?
MiniMax-M2.7 is the stronger model overall, scoring 37.4 to 24.1 on the Noometry Index.
Is Llama 2-70B or MiniMax-M2.7 better for coding?
MiniMax-M2.7 scores higher on coding benchmarks: 41.4 versus 30.8 in the Noometry coding category.
How many benchmarks do Llama 2-70B and MiniMax-M2.7 share?
18 benchmarks have published results for both models. Llama 2-70B has 35 scored results on Noometry and MiniMax-M2.7 has 30.