Model comparison
Claude 3 Haiku vs MiMo-V2.5
MiMo-V2.5 is the stronger model overall, scoring 43.4 to 25.9 on the Noometry Index.
Last verified . 18 shared benchmarks.
Summary
- They share 18 benchmarks with published results for both. Claude 3 Haiku scores higher in 0 categories and MiMo-V2.5 in 9 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in writing & preference, where MiMo-V2.5 leads 61.6 to 29.7.
- MiMo-V2.5 has downloadable open weights; the other is API-only.
Side by side
| Claude 3 Haiku | MiMo-V2.5 | |
|---|---|---|
| Provider | Anthropic | Xiaomi |
| Noometry Index | 25.9 | 43.4 |
| Released | 2024-03-07 | 2026-04-22 |
| Weights | Proprietary | Open |
| Context window | — | 1.05M |
| Max output | — | 131K |
| Input $ / M tokens | — | $0.14 |
| Output $ / M tokens | — | $0.28 |
| Results tracked | 37 | 23 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding MiMo-V2.5 leads
Claude 3 Haiku: 26.4 (#325), MiMo-V2.5: 43.9 (#81)
| Benchmark | Claude 3 Haiku | MiMo-V2.5 |
|---|---|---|
| LMArena Coding | 1199 | 1469 |
| LMArena WebDev | — | 1438 |
| SciCode | — | 43.1% |
| WeirdML | 9.8% | — |
| BigCodeBench Instruct | 39.4% | — |
| BigCodeBench Complete | 50.1% | — |
| CadEval | 12% | — |
| ALE-Bench | — | 513.95 |
| HumanEval+ | 68.9% | — |
| MBPP+ | 68.8% | — |
Reasoning MiMo-V2.5 leads
Claude 3 Haiku: 16.3 (#307), MiMo-V2.5: 28.6 (#101)
| Benchmark | Claude 3 Haiku | MiMo-V2.5 |
|---|---|---|
| LMArena Hard Prompts | 1174 | 1450 |
| Kagi LLM Benchmark | 34.2% | — |
| CritPt | — | 3.7% |
| DTBench | 50.1% | — |
| LMCA | 8.8% | — |
| Epoch Capabilities Index | 118.35 | — |
| ForecastBench | 53.2 | — |
| WinoGrande | 74.2% | — |
Math MiMo-V2.5 leads
Claude 3 Haiku: 9.8 (#319), MiMo-V2.5: 36.8 (#163)
| Benchmark | Claude 3 Haiku | MiMo-V2.5 |
|---|---|---|
| LMArena Math | 1188 | 1436 |
| OTIS Mock AIME 2024-2025 | 1.8% | — |
| ProofBench | — | 16% |
| MATH Level 5 | 14.9% | — |
Knowledge MiMo-V2.5 leads
Claude 3 Haiku: 17.3 (#285), MiMo-V2.5: 40.8 (#115)
| Benchmark | Claude 3 Haiku | MiMo-V2.5 |
|---|---|---|
| LMArena Expert | 1148 | 1460 |
| GPQA Diamond | 36.3% | — |
| Confabulations | 34.2% | — |
| MMLU | 73.8% | — |
Multimodal MiMo-V2.5 leads
Claude 3 Haiku: 23.6 (#128), MiMo-V2.5: 39.8 (#54)
| Benchmark | Claude 3 Haiku | MiMo-V2.5 |
|---|---|---|
| LMArena Vision | 950 | 1247 |
| ScienceQA | 72% | — |
Multilingual MiMo-V2.5 leads
Claude 3 Haiku: 36.0 (#243), MiMo-V2.5: 51.9 (#99)
| Benchmark | Claude 3 Haiku | MiMo-V2.5 |
|---|---|---|
| LMArena Non-English | 1178 | 1404 |
| LMArena Chinese | 1155 | 1468 |
| LMArena French | 1195 | 1447 |
| LMArena German | 1174 | 1421 |
| LMArena Japanese | 1102 | 1306 |
| LMArena Korean | 1109 | 1363 |
| LMArena Russian | 1204 | 1395 |
| LMArena Spanish | 1166 | 1416 |
Instruction Following MiMo-V2.5 leads
Claude 3 Haiku: 61.3 (#247), MiMo-V2.5: 75.5 (#60)
| Benchmark | Claude 3 Haiku | MiMo-V2.5 |
|---|---|---|
| LMArena Instruction Following | 1173 | 1434 |
Long Context MiMo-V2.5 leads
Claude 3 Haiku: 36.1 (#237), MiMo-V2.5: 44.2 (#73)
| Benchmark | Claude 3 Haiku | MiMo-V2.5 |
|---|---|---|
| LMArena Longer Query | 1190 | 1445 |
Writing & Preference MiMo-V2.5 leads
Claude 3 Haiku: 29.7 (#291), MiMo-V2.5: 61.6 (#86)
| Benchmark | Claude 3 Haiku | MiMo-V2.5 |
|---|---|---|
| LMArena Text | 1195 | 1428 |
| LMArena Creative Writing | 1157 | 1393 |
| LMArena Multi-Turn | 1190 | 1445 |
| EQ-Bench Creative Writing | 717 | — |
Frequently asked questions
Is Claude 3 Haiku better than MiMo-V2.5?
MiMo-V2.5 is the stronger model overall, scoring 43.4 to 25.9 on the Noometry Index.
Is Claude 3 Haiku or MiMo-V2.5 better for coding?
MiMo-V2.5 scores higher on coding benchmarks: 43.9 versus 26.4 in the Noometry coding category.
How many benchmarks do Claude 3 Haiku and MiMo-V2.5 share?
18 benchmarks have published results for both models. Claude 3 Haiku has 37 scored results on Noometry and MiMo-V2.5 has 23.