Model comparison
MiMo-V2.6-Pro vs Qwen3 235B-A22B
MiMo-V2.6-Pro is the stronger model overall, scoring 46.7 to 43.2 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 5 categories and Qwen3 235B-A22B in 3 categories; 7 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where MiMo-V2.6-Pro leads 30.4 to 15.8.
- MiMo-V2.6-Pro is cheaper at $0.43 / $0.87 per million input/output tokens, against $0.70 / $2.80 for Qwen3 235B-A22B.
- MiMo-V2.6-Pro accepts more context: 1.05M tokens versus 131K.
Side by side
| MiMo-V2.6-Pro | Qwen3 235B-A22B | |
|---|---|---|
| Provider | Xiaomi | Alibaba (Qwen) |
| Noometry Index | 46.7 | 43.2 |
| Released | 2026-09-22 | 2025-04 |
| Weights | Open | Open |
| Context window | 1.05M | 131K |
| Max output | 131K | 16K |
| Input $ / M tokens | $0.43 | $0.70 |
| Output $ / M tokens | $0.87 | $2.80 |
| Results tracked | 14 | 49 |
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: 53.2 (#27), Qwen3 235B-A22B: 43.8 (#72)
| Benchmark | MiMo-V2.6-Pro | Qwen3 235B-A22B |
|---|---|---|
| LMArena Coding | 1538 | 1445 |
| Aider Polyglot | — | 59.6% |
| LMArena WebDev | 1618 | — |
| SciCode | — | 42.4% |
| WeirdML | — | 41% |
Agentic & Tool Use Not comparable
MiMo-V2.6-Pro: —, Qwen3 235B-A22B: 33.7 (#48)
| Benchmark | MiMo-V2.6-Pro | Qwen3 235B-A22B |
|---|---|---|
| Berkeley Function Calling Leaderboard | — | 52.1% |
| Vending-Bench 2 | — | -11.34 |
Reasoning MiMo-V2.6-Pro leads
MiMo-V2.6-Pro: 30.4 (#76), Qwen3 235B-A22B: 15.8 (#302)
| Benchmark | MiMo-V2.6-Pro | Qwen3 235B-A22B |
|---|---|---|
| LMArena Hard Prompts | 1511 | 1433 |
| ARC-AGI-2 | — | 1.3% |
| SimpleBench | — | 31% |
| Kagi LLM Benchmark | — | 69.4% |
| ARC-AGI-1 | — | 11% |
| CritPt | — | 0% |
| Chess Puzzles | — | 12% |
| Mystery Game Puzzles | — | 9% |
| DTBench | — | 80.3% |
| LMCA | — | 29.3% |
| Epoch Capabilities Index | — | 143.87 |
| ForecastBench | — | 59.7 |
Math Qwen3 235B-A22B leads
MiMo-V2.6-Pro: 37.9 (#91), Qwen3 235B-A22B: 48.5 (#51)
| Benchmark | MiMo-V2.6-Pro | Qwen3 235B-A22B |
|---|---|---|
| LMArena Math | 1497 | 1431 |
| OTIS Mock AIME 2024-2025 | — | 86.7% |
| Omni-MATH | — | 71.8% |
| MATH Level 5 | — | 68.9% |
| FrontierMath (Feb 2025 set) | — | 8.5% |
| FrontierMath Tier 4 (v1) | — | 0% |
Knowledge Qwen3 235B-A22B leads
MiMo-V2.6-Pro: 43.0 (#85), Qwen3 235B-A22B: 49.5 (#67)
| Benchmark | MiMo-V2.6-Pro | Qwen3 235B-A22B |
|---|---|---|
| LMArena Expert | 1548 | 1460 |
| GPQA Diamond | — | 80.1% |
| SimpleQA Verified | — | 40.4% |
| MMLU-Pro | — | 84.4% |
| Confabulations | — | 15.6% |
| Vectara Hallucination Rate | — | 9.3% |
| GPQA (HELM) | — | 72.7% |
Multimodal Not comparable
MiMo-V2.6-Pro: 40.2 (#40), Qwen3 235B-A22B: —
| Benchmark | MiMo-V2.6-Pro | Qwen3 235B-A22B |
|---|---|---|
| LMArena Vision | 1265 | — |
Multilingual MiMo-V2.6-Pro leads
MiMo-V2.6-Pro: 57.0 (#15), Qwen3 235B-A22B: 52.5 (#84)
| Benchmark | MiMo-V2.6-Pro | Qwen3 235B-A22B |
|---|---|---|
| LMArena Non-English | 1474 | 1410 |
| LMArena Chinese | 1532 | 1481 |
| LMArena Russian | 1481 | 1411 |
| LMArena French | — | 1445 |
| LMArena German | — | 1432 |
| LMArena Japanese | — | 1400 |
| LMArena Korean | — | 1391 |
| LMArena Spanish | — | 1429 |
Instruction Following MiMo-V2.6-Pro leads
MiMo-V2.6-Pro: 78.2 (#10), Qwen3 235B-A22B: 72.8 (#132)
| Benchmark | MiMo-V2.6-Pro | Qwen3 235B-A22B |
|---|---|---|
| LMArena Instruction Following | 1490 | 1408 |
| IFEval | — | 83.5% |
Long Context Too close to call
MiMo-V2.6-Pro: 46.0 (#26), Qwen3 235B-A22B: 46.1 (#25)
| Benchmark | MiMo-V2.6-Pro | Qwen3 235B-A22B |
|---|---|---|
| LMArena Longer Query | 1500 | 1426 |
| Fiction.LiveBench | — | 75% |
Writing & Preference MiMo-V2.6-Pro leads
MiMo-V2.6-Pro: 66.9 (#32), Qwen3 235B-A22B: 59.8 (#103)
| Benchmark | MiMo-V2.6-Pro | Qwen3 235B-A22B |
|---|---|---|
| LMArena Text | 1491 | 1419 |
| LMArena Creative Writing | 1465 | 1384 |
| LMArena Multi-Turn | 1463 | 1432 |
| Short-Story Creative Writing | — | 83% |
| EQ-Bench Creative Writing | — | 1366 |
| WildBench | — | 86.6% |
Frequently asked questions
Is MiMo-V2.6-Pro better than Qwen3 235B-A22B?
MiMo-V2.6-Pro is the stronger model overall, scoring 46.7 to 43.2 on the Noometry Index.
Which is cheaper, MiMo-V2.6-Pro or Qwen3 235B-A22B?
MiMo-V2.6-Pro is cheaper. It lists at $0.43 per million input tokens and $0.87 per million output tokens; Qwen3 235B-A22B lists at $0.70 and $2.80.
Is MiMo-V2.6-Pro or Qwen3 235B-A22B better for coding?
MiMo-V2.6-Pro scores higher on coding benchmarks: 53.2 versus 43.8 in the Noometry coding category.
Which has the bigger context window?
MiMo-V2.6-Pro does, with 1.05M tokens against 131K.
How many benchmarks do MiMo-V2.6-Pro and Qwen3 235B-A22B share?
12 benchmarks have published results for both models. MiMo-V2.6-Pro has 14 scored results on Noometry and Qwen3 235B-A22B has 49.