MiMo-V2.6-Pro vs Muse Spark 1.2
MiMo-V2.6-Pro 
Muse Spark 1.2 Verdict
- MiMo-V2.6-Pro and Muse Spark 1.2 (xhigh) are level on quality (64.3 vs 63.4).
- MiMo-V2.6-Pro is stronger in composite, long context, reasoning.
- Muse Spark 1.2 (xhigh) is stronger in agents & tools, knowledge, coding, human preference, multimodal.
- MiMo-V2.6-Pro is 3.6× cheaper ($0.548 vs $2.00 per 1M blended).
- Muse Spark 1.2 (xhigh) streams 3.2× faster (173 vs 54 tokens per second).
| Metric | MiMo-V2.6-Pro | Muse Spark 1.2 (xhigh) |
|---|---|---|
| BenchLeader Index | 64.3 | 63.4 |
| Agents & tools score | 55.1 | 57.2 |
| Composite score | 84.8 | 67.8 |
| Knowledge score | 66.8 | 67.5 |
| Long context score | 69.1 | 65.3 |
| Reasoning score | 88.8 | 71.8 |
| Coding score | – | 56.6 |
| Human preference score | – | 70.6 |
| Multimodal score | – | 66.0 |
| Blended price $/M | $0.548 | $2.00 |
| Output speed | 54 tok/s | 173 tok/s |
| Time to first answer | 39.8 s | 27.9 s |
| Context window | 1.0M | 1.0M |
| SimpleQA Verified | – | 60.3% |
| SciCode | – | 56.4% |
| WeirdML | – | 60.3% |
| LMArena Text | – | 1500 |
| LMArena Hard Prompts | – | 1513 |
| LMArena Coding | – | 1536 |
| LMArena WebDev | – | 1534 |
| LMArena Vision | – | 1304 |
| LMArena Agent | – | -1.8 |
| LiveBench | – | 78.0% |
| LiveBench Reasoning | – | 90.0% |
| LiveBench Coding | – | 77.5% |
| LiveBench Agentic Coding | – | 57.6% |
| LiveBench Mathematics | – | 91.2% |
| LiveBench Data Analysis | – | 76.5% |
| LiveBench Language | – | 78.6% |
| LiveBench Instruction Following | – | 74.3% |
| AA Intelligence Index | 46.3 | 39.6 |
| AA-LCR | 86.3% | 79.0% |
| AA-Omniscience | 8.4 | 27.2 |
| GPQA Diamond (AA) | – | 90.4% |
| Humanity's Last Exam (AA) | 49.4% | 45.5% |
| SciCode (AA) | 60.9% | 57.4% |
| MMLU-Pro | – | 88.3% |
| IOI | – | 21.8% |
| LegalBench | – | 85.3% |
| CorpFin | – | 70.9% |
| TaxEval | – | 80.4% |
| Terminal-Bench 2.1 (Vals) | 67.8% | 69.7% |
| SWE-bench (Vals) | – | 86.6% |
| Vals Index | 59.7 | 57.0 |
| CritPt | 26.6% | 17.7% |
| GDPval (AA) | 58.7% | 49.1% |
| τ²-Bench Banking (AA) | – | 34.9% |
| BioMysteryBench | – | 64.8% |
| Code Migration | 43.0% | 29.9% |
| Excel Modeling Benchmark | 62.9% | 57.0% |
| Finance Agent v2 | 58.3% | 60.6% |
| Harvey's Legal Agent Benchmark | 10.8% | 25.4% |
| Legal Research Bench | 47.1% | 43.8% |
| MedCode | – | 49.4% |
| MedScribe | – | 90.1% |
| MMMU-Pro (Vals) | – | 86.1% |
| MortgageTax | – | 65.4% |
| Public Benefits Bench | – | 68.5% |
| SAGE | – | 47.7% |
| SkillsBench | – | 53.0% |
| Terminal-Bench 4.0 (Vals) | – | 5.6% |
| Terminal-Bench Science | 2.9% | – |
| Vibe Code Bench v1.1 | 85.2% | 79.1% |
| LMArena Creative Writing | – | 1452 |
| LMArena Instruction Following | – | 1480 |
| LMArena Multi-turn | – | 1520 |
| LMArena Longer Queries | – | 1502 |
| DeepSWE | – | 54.9% |
| LMCA | – | 48.4% |
| DTBench | – | 94.7% |
| GDP.pdf | – | 16.0% |
| FrontierSWE | – | 12.0% |
| Terminal-Bench 4.0 (AA) | 34.9% | 7.1% |
| Terminal-Bench 2.1 (AA) | – | 80.2% |
| AutomationBench | 58.6% | – |
| GDP.pdf | 19.2% | – |
| AA-Omniscience: accuracy | 34.9% | 45.4% |
| AA-Omniscience: non-hallucination | 59.4% | 66.7% |
| AA-Briefcase | 1522 | – |
Data as of 2026-09-23. Best configuration of each model; every score links to its source on the model pages.
MiMo-V2.6-Pro vs Muse Spark 1.2: questions
- Is MiMo-V2.6-Pro better than Muse Spark 1.2?
- MiMo-V2.6-Pro and Muse Spark 1.2 (xhigh) are level on quality (64.3 vs 63.4). The BenchLeader Index combines every independent quality benchmark; MiMo-V2.6-Pro is ahead overall as of 2026-09-23, but check the category scores for your use.
- Is MiMo-V2.6-Pro better than Muse Spark 1.2 for agentic tasks?
- Muse Spark 1.2 scores higher in agentic tasks (57 vs 55 on the category index, where 50 is average).
- Which is cheaper, MiMo-V2.6-Pro or Muse Spark 1.2?
- MiMo-V2.6-Pro is cheaper: $0.548 against $2.00 per million tokens, blended at three input tokens per output token.
- Which is faster, MiMo-V2.6-Pro or Muse Spark 1.2?
- Muse Spark 1.2 streams faster: 173 against 54 output tokens per second.
- Which has the larger context window?
- Both accept 1.0M tokens of context.