Model comparison
Claude Haiku 4.5 vs MiMo-V2.5
MiMo-V2.5 is the stronger model overall, scoring 43.0 to 39.3 on the Noometry Index.
Last verified . 21 shared benchmarks.
Summary
- They share 21 benchmarks with published results for both. Claude Haiku 4.5 scores higher in 2 categories and MiMo-V2.5 in 7 categories; 7 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where MiMo-V2.5 leads 27.8 to 15.1.
- MiMo-V2.5 is cheaper at $0.14 / $0.28 per million input/output tokens, against $1 / $5 for Claude Haiku 4.5.
- MiMo-V2.5 accepts more context: 1.05M tokens versus 200K.
- MiMo-V2.5 has downloadable open weights; the other is API-only.
Side by side
| Claude Haiku 4.5 | MiMo-V2.5 | |
|---|---|---|
| Provider | Anthropic | Xiaomi |
| Noometry Index | 39.3 | 43.0 |
| Released | 2025-10-15 | 2026-04-22 |
| Weights | Proprietary | Open |
| Context window | 200K | 1.05M |
| Max output | 64K | 131K |
| Input $ / M tokens | $1 | $0.14 |
| Output $ / M tokens | $5 | $0.28 |
| Results tracked | 53 | 23 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Too close to call
Claude Haiku 4.5: 43.9 (#71), MiMo-V2.5: 43.6 (#75)
| Benchmark | Claude Haiku 4.5 | MiMo-V2.5 |
|---|---|---|
| LMArena WebDev | 1330 | 1438 |
| SciCode | 43.3% | 43.1% |
| LMArena Coding | 1454 | 1469 |
| ALE-Bench | 653.48 | 513.95 |
| SWE-bench Verified (bash only) | 66.6% | — |
| SWE-bench Multilingual | 64.7% | — |
| WeirdML | 45.4% | — |
Agentic & Tool Use Not comparable
Claude Haiku 4.5: 33.5 (#50), MiMo-V2.5: —
| Benchmark | Claude Haiku 4.5 | MiMo-V2.5 |
|---|---|---|
| Terminal-Bench | 35.5% | — |
| Berkeley Function Calling Leaderboard | 68.7% | — |
| DeepResearch Bench | 45.5% | — |
| BALROG | 31.2% | — |
| ExploitBench | 13.7% | — |
| Vending-Bench 2 | 458.89 | — |
Reasoning MiMo-V2.5 leads
Claude Haiku 4.5: 15.1 (#310), MiMo-V2.5: 27.8 (#98)
| Benchmark | Claude Haiku 4.5 | MiMo-V2.5 |
|---|---|---|
| CritPt | 0% | 3.7% |
| LMArena Hard Prompts | 1420 | 1450 |
| ARC-AGI-2 | 4% | — |
| NYT Connections (extended) | 14.3% | — |
| ARC-AGI-1 | 47.7% | — |
| Chess Puzzles | 8% | — |
| DTBench | 73.6% | — |
| LMCA | 30.9% | — |
| Epoch Capabilities Index | 142.42 | — |
| ForecastBench | 61.4 | — |
Math Claude Haiku 4.5 leads
Claude Haiku 4.5: 43.0 (#69), MiMo-V2.5: 35.4 (#145)
| Benchmark | Claude Haiku 4.5 | MiMo-V2.5 |
|---|---|---|
| LMArena Math | 1395 | 1436 |
| OTIS Mock AIME 2024-2025 | 66.7% | — |
| ProofBench | — | 16% |
| Omni-MATH | 56.1% | — |
| MATH Level 5 | 96.4% | — |
| FrontierMath (Feb 2025 set) | 5.9% | — |
| FrontierMath Tier 4 (v1) | 2.1% | — |
Knowledge MiMo-V2.5 leads
Claude Haiku 4.5: 37.6 (#146), MiMo-V2.5: 40.2 (#109)
| Benchmark | Claude Haiku 4.5 | MiMo-V2.5 |
|---|---|---|
| LMArena Expert | 1443 | 1459 |
| GPQA Diamond | 71.2% | — |
| SimpleQA Verified | 13.2% | — |
| MMLU-Pro | 77.7% | — |
| Vectara Hallucination Rate | 9.8% | — |
| GPQA (HELM) | 60.5% | — |
Multimodal MiMo-V2.5 leads
Claude Haiku 4.5: 27.0 (#111), MiMo-V2.5: 39.0 (#53)
| Benchmark | Claude Haiku 4.5 | MiMo-V2.5 |
|---|---|---|
| LMArena Vision | — | 1245 |
| Blueprint-Bench 2 | 0% | — |
| LMArena Document | 1420 | — |
Multilingual MiMo-V2.5 leads
Claude Haiku 4.5: 50.1 (#124), MiMo-V2.5: 52.0 (#96)
| Benchmark | Claude Haiku 4.5 | MiMo-V2.5 |
|---|---|---|
| LMArena Non-English | 1376 | 1404 |
| LMArena Chinese | 1415 | 1467 |
| LMArena French | 1408 | 1447 |
| LMArena German | 1377 | 1421 |
| LMArena Japanese | 1339 | 1304 |
| LMArena Korean | 1346 | 1363 |
| LMArena Russian | 1380 | 1395 |
| LMArena Spanish | 1420 | 1412 |
Instruction Following MiMo-V2.5 leads
Claude Haiku 4.5: 71.6 (#144), MiMo-V2.5: 75.7 (#55)
| Benchmark | Claude Haiku 4.5 | MiMo-V2.5 |
|---|---|---|
| LMArena Instruction Following | 1414 | 1433 |
| IFEval | 80.1% | — |
Long Context Too close to call
Claude Haiku 4.5: 43.6 (#86), MiMo-V2.5: 44.1 (#69)
| Benchmark | Claude Haiku 4.5 | MiMo-V2.5 |
|---|---|---|
| LMArena Longer Query | 1427 | 1444 |
Writing & Preference MiMo-V2.5 leads
Claude Haiku 4.5: 58.1 (#119), MiMo-V2.5: 61.8 (#82)
| Benchmark | Claude Haiku 4.5 | MiMo-V2.5 |
|---|---|---|
| LMArena Text | 1396 | 1427 |
| LMArena Creative Writing | 1371 | 1392 |
| LMArena Multi-Turn | 1409 | 1445 |
| WildBench | 83.9% | — |
| EQ-Bench 4 | 1064 | — |
Frequently asked questions
Is Claude Haiku 4.5 better than MiMo-V2.5?
MiMo-V2.5 is the stronger model overall, scoring 43.0 to 39.3 on the Noometry Index.
Which is cheaper, Claude Haiku 4.5 or MiMo-V2.5?
MiMo-V2.5 is cheaper. It lists at $0.14 per million input tokens and $0.28 per million output tokens; Claude Haiku 4.5 lists at $1 and $5.
Is Claude Haiku 4.5 or MiMo-V2.5 better for coding?
They score almost the same on coding (43.9 vs 43.6); test both on your own repository before choosing.
Which has the bigger context window?
MiMo-V2.5 does, with 1.05M tokens against 200K.
How many benchmarks do Claude Haiku 4.5 and MiMo-V2.5 share?
21 benchmarks have published results for both models. Claude Haiku 4.5 has 53 scored results on Noometry and MiMo-V2.5 has 23.