Model comparison
MiMo-V2.6-Pro vs Qwen2.5-Coder-32B
MiMo-V2.6-Pro is the stronger model overall, scoring 50.3 to 33.4 on the Noometry Index.
Last verified . 12 shared benchmarks.
Alibaba (Qwen)Summary
- They share 12 benchmarks with published results for both. MiMo-V2.6-Pro scores higher in 8 categories and Qwen2.5-Coder-32B in 0 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in coding, where MiMo-V2.6-Pro leads 55.5 to 22.6.
- MiMo-V2.6-Pro is cheaper at $0.43 / $0.87 per million input/output tokens, against $0.66 / $1 for Qwen2.5-Coder-32B.
- MiMo-V2.6-Pro accepts more context: 1.05M tokens versus 33K.
Side by side
| MiMo-V2.6-Pro | Qwen2.5-Coder-32B | |
|---|---|---|
| Provider | Xiaomi | Alibaba (Qwen) |
| Noometry Index | 50.3 | 33.4 |
| Released | 2026-09-21 | 2024-09-18 |
| Weights | Open | Open |
| Context window | 1.05M | 33K |
| Max output | 131K | 29K |
| Input $ / M tokens | $0.43 | $0.66 |
| Output $ / M tokens | $0.87 | $1 |
| Results tracked | 19 | 31 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding MiMo-V2.6-Pro leads
MiMo-V2.6-Pro: 55.5 (#23), Qwen2.5-Coder-32B: 22.6 (#333)
| Benchmark | MiMo-V2.6-Pro | Qwen2.5-Coder-32B |
|---|---|---|
| LMArena Coding | 1534 | 1276 |
| SWE-bench Verified (bash only) | — | 9% |
| Aider Polyglot | — | 16.4% |
| LMArena WebDev | 1629 | — |
| SciCode | 60.9% | — |
| BigCodeBench Instruct | — | 49% |
| LiveBench Coding | — | 56.9% |
| BigCodeBench Complete | — | 58% |
| ALE-Bench | 1,158 | — |
| HumanEval+ | — | 87.2% |
| MBPP+ | — | 77% |
Agentic & Tool Use Not comparable
MiMo-V2.6-Pro: 37.5 (#35), Qwen2.5-Coder-32B: —
| Benchmark | MiMo-V2.6-Pro | Qwen2.5-Coder-32B |
|---|---|---|
| APEX-Agents | 59.5% | — |
Reasoning MiMo-V2.6-Pro leads
MiMo-V2.6-Pro: 43.1 (#50), Qwen2.5-Coder-32B: 21.2 (#225)
| Benchmark | MiMo-V2.6-Pro | Qwen2.5-Coder-32B |
|---|---|---|
| LMArena Hard Prompts | 1512 | 1251 |
| CritPt | 26.6% | — |
| LiveBench Reasoning | — | 42.1% |
| LiveBench Data Analysis | — | 49.9% |
| Epoch Capabilities Index | — | 119.49 |
| HellaSwag | — | 83% |
| LiveBench | — | 46.2% |
| WinoGrande | — | 80.8% |
Math MiMo-V2.6-Pro leads
MiMo-V2.6-Pro: 54.5 (#45), Qwen2.5-Coder-32B: 33.3 (#204)
| Benchmark | MiMo-V2.6-Pro | Qwen2.5-Coder-32B |
|---|---|---|
| LMArena Math | 1494 | 1251 |
| ProofBench | 70% | — |
| LiveBench Math | — | 46.6% |
| GSM8K | — | 93% |
Knowledge MiMo-V2.6-Pro leads
MiMo-V2.6-Pro: 43.5 (#92), Qwen2.5-Coder-32B: 33.4 (#203)
| Benchmark | MiMo-V2.6-Pro | Qwen2.5-Coder-32B |
|---|---|---|
| LMArena Expert | 1543 | 1221 |
| ARC (AI2) Challenge | — | 70.5% |
| MMLU | — | 79.1% |
Multimodal Not comparable
MiMo-V2.6-Pro: 40.8 (#43), Qwen2.5-Coder-32B: —
| Benchmark | MiMo-V2.6-Pro | Qwen2.5-Coder-32B |
|---|---|---|
| LMArena Vision | 1264 | — |
Multilingual MiMo-V2.6-Pro leads
MiMo-V2.6-Pro: 56.9 (#14), Qwen2.5-Coder-32B: 37.8 (#235)
| Benchmark | MiMo-V2.6-Pro | Qwen2.5-Coder-32B |
|---|---|---|
| LMArena Non-English | 1474 | 1205 |
| LMArena Chinese | 1529 | 1222 |
| LMArena Russian | 1480 | 1228 |
Instruction Following MiMo-V2.6-Pro leads
MiMo-V2.6-Pro: 78.2 (#12), Qwen2.5-Coder-32B: 61.4 (#245)
| Benchmark | MiMo-V2.6-Pro | Qwen2.5-Coder-32B |
|---|---|---|
| LMArena Instruction Following | 1493 | 1223 |
| LiveBench Instruction Following | — | 58.7% |
Long Context MiMo-V2.6-Pro leads
MiMo-V2.6-Pro: 46.0 (#27), Qwen2.5-Coder-32B: 38.0 (#208)
| Benchmark | MiMo-V2.6-Pro | Qwen2.5-Coder-32B |
|---|---|---|
| LMArena Longer Query | 1501 | 1251 |
Writing & Preference MiMo-V2.6-Pro leads
MiMo-V2.6-Pro: 66.8 (#33), Qwen2.5-Coder-32B: 41.6 (#240)
| Benchmark | MiMo-V2.6-Pro | Qwen2.5-Coder-32B |
|---|---|---|
| LMArena Text | 1492 | 1230 |
| LMArena Creative Writing | 1468 | 1174 |
| LMArena Multi-Turn | 1464 | 1222 |
| LiveBench Language | — | 23.3% |
Frequently asked questions
Is MiMo-V2.6-Pro better than Qwen2.5-Coder-32B?
MiMo-V2.6-Pro is the stronger model overall, scoring 50.3 to 33.4 on the Noometry Index.
Which is cheaper, MiMo-V2.6-Pro or Qwen2.5-Coder-32B?
MiMo-V2.6-Pro is cheaper. It lists at $0.43 per million input tokens and $0.87 per million output tokens; Qwen2.5-Coder-32B lists at $0.66 and $1.
Is MiMo-V2.6-Pro or Qwen2.5-Coder-32B better for coding?
MiMo-V2.6-Pro scores higher on coding benchmarks: 55.5 versus 22.6 in the Noometry coding category.
Which has the bigger context window?
MiMo-V2.6-Pro does, with 1.05M tokens against 33K.
How many benchmarks do MiMo-V2.6-Pro and Qwen2.5-Coder-32B share?
12 benchmarks have published results for both models. MiMo-V2.6-Pro has 19 scored results on Noometry and Qwen2.5-Coder-32B has 31.