Model comparison
Llama 3-70B vs MiMo-V2.5-Pro
MiMo-V2.5-Pro is the stronger model overall, scoring 45.2 to 28.8 on the Noometry Index.
Last verified . 18 shared benchmarks.
MetaSummary
- They share 18 benchmarks with published results for both. Llama 3-70B scores higher in 0 categories and MiMo-V2.5-Pro in 8 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in math, where MiMo-V2.5-Pro leads 40.0 to 12.8.
- The biggest single-benchmark swing is DTBench: 54.2% for Llama 3-70B and 84.5% for MiMo-V2.5-Pro.
Side by side
| Llama 3-70B | MiMo-V2.5-Pro | |
|---|---|---|
| Provider | Meta | Xiaomi |
| Noometry Index | 28.8 | 45.2 |
| Released | 2024-04-18 | 2026-04-22 |
| Weights | Open | Open |
| Context window | — | 1.05M |
| Max output | — | 131K |
| Input $ / M tokens | — | $0.43 |
| Output $ / M tokens | — | $0.87 |
| Results tracked | 31 | 27 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding MiMo-V2.5-Pro leads
Llama 3-70B: 35.8 (#218), MiMo-V2.5-Pro: 47.4 (#60)
| Benchmark | Llama 3-70B | MiMo-V2.5-Pro |
|---|---|---|
| LMArena Coding | 1206 | 1503 |
| LMArena WebDev | — | 1479 |
| SciCode | — | 50.2% |
| BigCodeBench Instruct | 43.6% | — |
| BigCodeBench Complete | 54.5% | — |
| ALE-Bench | — | 899.8 |
| HumanEval+ | 72% | — |
| MBPP+ | 69% | — |
Agentic & Tool Use Not comparable
Llama 3-70B: 21.1 (#139), MiMo-V2.5-Pro: —
| Benchmark | Llama 3-70B | MiMo-V2.5-Pro |
|---|---|---|
| Cybench | 5% | — |
Reasoning MiMo-V2.5-Pro leads
Llama 3-70B: 18.0 (#288), MiMo-V2.5-Pro: 26.8 (#130)
| Benchmark | Llama 3-70B | MiMo-V2.5-Pro |
|---|---|---|
| LMArena Hard Prompts | 1195 | 1488 |
| DTBench | 54.2% | 84.5% |
| Kagi LLM Benchmark | 35.1% | — |
| NYT Connections (extended) | — | 34.4% |
| CritPt | — | 4% |
| LMCA | — | 29.5% |
| Epoch Capabilities Index | 122.93 | — |
| ForecastBench | 57.1 | — |
| WinoGrande | 83.5% | — |
Math MiMo-V2.5-Pro leads
Llama 3-70B: 12.8 (#305), MiMo-V2.5-Pro: 40.0 (#96)
| Benchmark | Llama 3-70B | MiMo-V2.5-Pro |
|---|---|---|
| LMArena Math | 1218 | 1481 |
| OTIS Mock AIME 2024-2025 | 4.3% | — |
| ProofBench | — | 22% |
| MATH Level 5 | 22.6% | — |
Knowledge MiMo-V2.5-Pro leads
Llama 3-70B: 20.8 (#277), MiMo-V2.5-Pro: 42.2 (#98)
| Benchmark | Llama 3-70B | MiMo-V2.5-Pro |
|---|---|---|
| LMArena Expert | 1149 | 1503 |
| GPQA Diamond | 40.6% | — |
| MMLU | 79.3% | — |
Multilingual MiMo-V2.5-Pro leads
Llama 3-70B: 33.6 (#251), MiMo-V2.5-Pro: 55.1 (#34)
| Benchmark | Llama 3-70B | MiMo-V2.5-Pro |
|---|---|---|
| LMArena Non-English | 1142 | 1449 |
| LMArena Chinese | 1114 | 1507 |
| LMArena French | 1232 | 1488 |
| LMArena German | 1169 | 1458 |
| LMArena Japanese | 1017 | 1412 |
| LMArena Korean | 1017 | 1437 |
| LMArena Russian | 1159 | 1450 |
| LMArena Spanish | 1241 | 1471 |
Instruction Following MiMo-V2.5-Pro leads
Llama 3-70B: 62.5 (#238), MiMo-V2.5-Pro: 77.5 (#21)
| Benchmark | Llama 3-70B | MiMo-V2.5-Pro |
|---|---|---|
| LMArena Instruction Following | 1194 | 1477 |
Long Context MiMo-V2.5-Pro leads
Llama 3-70B: 35.6 (#240), MiMo-V2.5-Pro: 45.4 (#37)
| Benchmark | Llama 3-70B | MiMo-V2.5-Pro |
|---|---|---|
| LMArena Longer Query | 1174 | 1483 |
Writing & Preference MiMo-V2.5-Pro leads
Llama 3-70B: 42.8 (#231), MiMo-V2.5-Pro: 65.3 (#49)
| Benchmark | Llama 3-70B | MiMo-V2.5-Pro |
|---|---|---|
| LMArena Text | 1221 | 1465 |
| LMArena Creative Writing | 1210 | 1440 |
| LMArena Multi-Turn | 1223 | 1477 |
| EQ-Bench Creative Writing | — | 1493 |
| EQ-Bench 4 | — | 1208 |
Frequently asked questions
Is Llama 3-70B better than MiMo-V2.5-Pro?
MiMo-V2.5-Pro is the stronger model overall, scoring 45.2 to 28.8 on the Noometry Index.
Is Llama 3-70B or MiMo-V2.5-Pro better for coding?
MiMo-V2.5-Pro scores higher on coding benchmarks: 47.4 versus 35.8 in the Noometry coding category.
How many benchmarks do Llama 3-70B and MiMo-V2.5-Pro share?
18 benchmarks have published results for both models. Llama 3-70B has 31 scored results on Noometry and MiMo-V2.5-Pro has 27.