Model comparison
MiMo-V2.5 vs Qwen3.5 27B
MiMo-V2.5 is the stronger model overall, scoring 43.0 to 41.4 on the Noometry Index.
Last verified . 20 shared benchmarks.
Alibaba (Qwen)Summary
- They share 20 benchmarks with published results for both. MiMo-V2.5 scores higher in 8 categories and Qwen3.5 27B in 1 category; 8 gaps are clear of the uncertainty.
- The widest gap is in coding, where MiMo-V2.5 leads 43.6 to 38.6.
- MiMo-V2.5 is cheaper at $0.14 / $0.28 per million input/output tokens, against $0.30 / $2.40 for Qwen3.5 27B.
- MiMo-V2.5 accepts more context: 1.05M tokens versus 262K.
Side by side
| MiMo-V2.5 | Qwen3.5 27B | |
|---|---|---|
| Provider | Xiaomi | Alibaba (Qwen) |
| Noometry Index | 43.0 | 41.4 |
| Released | 2026-04-22 | 2026-02-23 |
| Weights | Open | Open |
| Context window | 1.05M | 262K |
| Max output | 131K | 66K |
| Input $ / M tokens | $0.14 | $0.30 |
| Output $ / M tokens | $0.28 | $2.40 |
| Results tracked | 23 | 28 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding MiMo-V2.5 leads
MiMo-V2.5: 43.6 (#75), Qwen3.5 27B: 38.6 (#159)
| Benchmark | MiMo-V2.5 | Qwen3.5 27B |
|---|---|---|
| LMArena WebDev | 1438 | 1357 |
| LMArena Coding | 1469 | 1427 |
| ALE-Bench | 513.95 | 349.45 |
| SciCode | 43.1% | — |
| WeirdML | — | 39.5% |
Agentic & Tool Use Not comparable
MiMo-V2.5: —, Qwen3.5 27B: —
| Benchmark | MiMo-V2.5 | Qwen3.5 27B |
|---|---|---|
| Vending-Bench 2 | — | 201.98 |
Reasoning MiMo-V2.5 leads
MiMo-V2.5: 27.8 (#98), Qwen3.5 27B: 26.6 (#116)
| Benchmark | MiMo-V2.5 | Qwen3.5 27B |
|---|---|---|
| LMArena Hard Prompts | 1450 | 1414 |
| NYT Connections (extended) | — | 47.9% |
| CritPt | 3.7% | — |
| Thematic Generalization | — | 45.5% |
| DTBench | — | 82.4% |
| LMCA | — | 34% |
Math Qwen3.5 27B leads
MiMo-V2.5: 35.4 (#145), Qwen3.5 27B: 36.8 (#103)
| Benchmark | MiMo-V2.5 | Qwen3.5 27B |
|---|---|---|
| LMArena Math | 1436 | 1428 |
| MathArena Final-Answer Competitions | — | 56.7% |
| ProofBench | 16% | — |
Knowledge MiMo-V2.5 leads
MiMo-V2.5: 40.2 (#109), Qwen3.5 27B: 37.4 (#149)
| Benchmark | MiMo-V2.5 | Qwen3.5 27B |
|---|---|---|
| LMArena Expert | 1459 | 1428 |
| Vectara Hallucination Rate | — | 12.1% |
Multimodal Too close to call
MiMo-V2.5: 39.0 (#53), Qwen3.5 27B: 38.7 (#56)
| Benchmark | MiMo-V2.5 | Qwen3.5 27B |
|---|---|---|
| LMArena Vision | 1245 | 1240 |
Multilingual MiMo-V2.5 leads
MiMo-V2.5: 52.0 (#96), Qwen3.5 27B: 51.0 (#112)
| Benchmark | MiMo-V2.5 | Qwen3.5 27B |
|---|---|---|
| LMArena Non-English | 1404 | 1390 |
| LMArena Chinese | 1467 | 1478 |
| LMArena French | 1447 | 1409 |
| LMArena German | 1421 | 1393 |
| LMArena Japanese | 1304 | 1345 |
| LMArena Korean | 1363 | 1359 |
| LMArena Russian | 1395 | 1389 |
| LMArena Spanish | 1412 | 1408 |
Instruction Following MiMo-V2.5 leads
MiMo-V2.5: 75.7 (#55), Qwen3.5 27B: 73.7 (#114)
| Benchmark | MiMo-V2.5 | Qwen3.5 27B |
|---|---|---|
| LMArena Instruction Following | 1433 | 1393 |
Long Context MiMo-V2.5 leads
MiMo-V2.5: 44.1 (#69), Qwen3.5 27B: 43.1 (#103)
| Benchmark | MiMo-V2.5 | Qwen3.5 27B |
|---|---|---|
| LMArena Longer Query | 1444 | 1413 |
Writing & Preference MiMo-V2.5 leads
MiMo-V2.5: 61.8 (#82), Qwen3.5 27B: 59.5 (#106)
| Benchmark | MiMo-V2.5 | Qwen3.5 27B |
|---|---|---|
| LMArena Text | 1427 | 1409 |
| LMArena Creative Writing | 1392 | 1362 |
| LMArena Multi-Turn | 1445 | 1410 |
Frequently asked questions
Is MiMo-V2.5 better than Qwen3.5 27B?
MiMo-V2.5 is the stronger model overall, scoring 43.0 to 41.4 on the Noometry Index.
Which is cheaper, MiMo-V2.5 or Qwen3.5 27B?
MiMo-V2.5 is cheaper. It lists at $0.14 per million input tokens and $0.28 per million output tokens; Qwen3.5 27B lists at $0.30 and $2.40.
Is MiMo-V2.5 or Qwen3.5 27B better for coding?
MiMo-V2.5 scores higher on coding benchmarks: 43.6 versus 38.6 in the Noometry coding category.
Which has the bigger context window?
MiMo-V2.5 does, with 1.05M tokens against 262K.
How many benchmarks do MiMo-V2.5 and Qwen3.5 27B share?
20 benchmarks have published results for both models. MiMo-V2.5 has 23 scored results on Noometry and Qwen3.5 27B has 28.