Model comparison
MiMo-V2.5 vs Qwen3.5 397B-A17B
Qwen3.5 397B-A17B is the stronger model overall, scoring 45.7 to 43.0 on the Noometry Index. MiMo-V2.5 costs 7.7× less per token, which makes it the better buy when Qwen3.5 397B-A17B's lead doesn't matter for your workload.
Last verified . 19 shared benchmarks.
Alibaba (Qwen)Summary
- They share 19 benchmarks with published results for both. MiMo-V2.5 scores higher in 3 categories and Qwen3.5 397B-A17B in 6 categories; 6 gaps are clear of the uncertainty.
- The widest gap is in knowledge, where Qwen3.5 397B-A17B leads 53.3 to 40.2.
- MiMo-V2.5 is cheaper at $0.14 / $0.28 per million input/output tokens, against $0.60 / $3.60 for Qwen3.5 397B-A17B.
- MiMo-V2.5 accepts more context: 1.05M tokens versus 262K.
Side by side
| MiMo-V2.5 | Qwen3.5 397B-A17B | |
|---|---|---|
| Provider | Xiaomi | Alibaba (Qwen) |
| Noometry Index | 43.0 | 45.7 |
| Released | 2026-04-22 | 2026-02-01 |
| Weights | Open | Open |
| Context window | 1.05M | 262K |
| Max output | 131K | 66K |
| Input $ / M tokens | $0.14 | $0.60 |
| Output $ / M tokens | $0.28 | $3.60 |
| Results tracked | 23 | 36 |
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 397B-A17B: 41.7 (#108)
| Benchmark | MiMo-V2.5 | Qwen3.5 397B-A17B |
|---|---|---|
| LMArena WebDev | 1438 | 1400 |
| LMArena Coding | 1469 | 1464 |
| SciCode | 43.1% | — |
| ALE-Bench | 513.95 | — |
Agentic & Tool Use Not comparable
MiMo-V2.5: —, Qwen3.5 397B-A17B: 33.2 (#51)
| Benchmark | MiMo-V2.5 | Qwen3.5 397B-A17B |
|---|---|---|
| APEX-Agents | — | 24.9% |
| τ²-bench Airline | — | 81.5% |
| τ²-bench Banking | — | 9.8% |
| τ²-bench Retail | — | 84.4% |
| τ²-bench Telecom | — | 97.8% |
Reasoning Qwen3.5 397B-A17B leads
MiMo-V2.5: 27.8 (#98), Qwen3.5 397B-A17B: 33.6 (#62)
| Benchmark | MiMo-V2.5 | Qwen3.5 397B-A17B |
|---|---|---|
| LMArena Hard Prompts | 1450 | 1448 |
| Kagi LLM Benchmark | — | 73.7% |
| NYT Connections (extended) | — | 58.9% |
| CritPt | 3.7% | — |
| Chess Puzzles | — | 13% |
| Thematic Generalization | — | 65.1% |
| Mystery Game Puzzles | — | 18% |
| DTBench | — | 87.5% |
| LMCA | — | 37.9% |
| Epoch Capabilities Index | — | 146.64 |
Math Qwen3.5 397B-A17B leads
MiMo-V2.5: 35.4 (#145), Qwen3.5 397B-A17B: 44.9 (#64)
| Benchmark | MiMo-V2.5 | Qwen3.5 397B-A17B |
|---|---|---|
| LMArena Math | 1436 | 1455 |
| FrontierMath (Tiers 1-3) | — | 31.2% |
| OTIS Mock AIME 2024-2025 | — | 88.9% |
| ProofBench | 16% | — |
Knowledge Qwen3.5 397B-A17B leads
MiMo-V2.5: 40.2 (#109), Qwen3.5 397B-A17B: 53.3 (#53)
| Benchmark | MiMo-V2.5 | Qwen3.5 397B-A17B |
|---|---|---|
| LMArena Expert | 1459 | 1461 |
| GPQA Diamond | — | 86.4% |
Multimodal Qwen3.5 397B-A17B leads
MiMo-V2.5: 39.0 (#53), Qwen3.5 397B-A17B: 40.0 (#42)
| Benchmark | MiMo-V2.5 | Qwen3.5 397B-A17B |
|---|---|---|
| LMArena Vision | 1245 | 1262 |
Multilingual Qwen3.5 397B-A17B leads
MiMo-V2.5: 52.0 (#96), Qwen3.5 397B-A17B: 54.0 (#56)
| Benchmark | MiMo-V2.5 | Qwen3.5 397B-A17B |
|---|---|---|
| LMArena Non-English | 1404 | 1431 |
| LMArena Chinese | 1467 | 1501 |
| LMArena French | 1447 | 1461 |
| LMArena German | 1421 | 1447 |
| LMArena Japanese | 1304 | 1427 |
| LMArena Korean | 1363 | 1382 |
| LMArena Russian | 1395 | 1429 |
| LMArena Spanish | 1412 | 1443 |
Instruction Following Too close to call
MiMo-V2.5: 75.7 (#55), Qwen3.5 397B-A17B: 75.2 (#73)
| Benchmark | MiMo-V2.5 | Qwen3.5 397B-A17B |
|---|---|---|
| LMArena Instruction Following | 1433 | 1424 |
Long Context Too close to call
MiMo-V2.5: 44.1 (#69), Qwen3.5 397B-A17B: 44.1 (#71)
| Benchmark | MiMo-V2.5 | Qwen3.5 397B-A17B |
|---|---|---|
| LMArena Longer Query | 1444 | 1442 |
Writing & Preference Too close to call
MiMo-V2.5: 61.8 (#82), Qwen3.5 397B-A17B: 62.5 (#74)
| Benchmark | MiMo-V2.5 | Qwen3.5 397B-A17B |
|---|---|---|
| LMArena Text | 1427 | 1438 |
| LMArena Creative Writing | 1392 | 1401 |
| LMArena Multi-Turn | 1445 | 1447 |
| EQ-Bench Creative Writing | — | 1478 |
Frequently asked questions
Is MiMo-V2.5 better than Qwen3.5 397B-A17B?
Qwen3.5 397B-A17B is the stronger model overall, scoring 45.7 to 43.0 on the Noometry Index. MiMo-V2.5 costs 7.7× less per token, which makes it the better buy when Qwen3.5 397B-A17B's lead doesn't matter for your workload.
Which is cheaper, MiMo-V2.5 or Qwen3.5 397B-A17B?
MiMo-V2.5 is cheaper. It lists at $0.14 per million input tokens and $0.28 per million output tokens; Qwen3.5 397B-A17B lists at $0.60 and $3.60.
Is MiMo-V2.5 or Qwen3.5 397B-A17B better for coding?
MiMo-V2.5 scores higher on coding benchmarks: 43.6 versus 41.7 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 397B-A17B share?
19 benchmarks have published results for both models. MiMo-V2.5 has 23 scored results on Noometry and Qwen3.5 397B-A17B has 36.