Up to four models
Claude Opus 5.5 vs MiMo-V2.6-Flash
On the index (GPT-4 = 100), Claude Opus 5.5 scores 195 and MiMo-V2.6-Flash 168. Side by side on the same third-party tests, with price and context. The link in your address bar shares this exact comparison.
Claude Opus 5.5
195178–211
WeightsClosed
ReleasedSep 2026
Price in / out$4.0 / $20
Context1M
Core tests taken6 of 9
MiMo-V2.6-Flash
168149–186
WeightsOpen
ReleasedSep 2026
Price in / out$0.14 / $0.28
Context1.05M
Core tests taken2 of 9
Claude Opus 5.5 leads MiMo-V2.6-Flash by 27 points, ahead on 2 of 2 shared core tests, at 46× the price.
Test by testthird-party results
Claude Opus 5.5MiMo-V2.6-Flash
Every result as a table
| Test | Claude Opus 5.5 | MiMo-V2.6-Flash |
|---|---|---|
| Maths | ||
| ProofBench v1.1 | 100.0% | 63.0% |
| Coding | ||
| Terminal-Bench 4.0 | 65.1% | 24.2% |
| Code Migration | 66.6% | 40.9% |
| CyberBench v1.1 | 74.6% | 75.4% |
| IOI | 95.1% | 47.7% |
| ProgramBench | 18.5% | 0.5% |
| Vibe Code Bench v1.1 | 90.3% | 79.0% |
| Agents | ||
| BioMysteryBench | 79.3% | 69.3% |
| SRE Bench | 33.6% | 4.2% |
| Terminal-Bench-Science | 47.1% | 4.3% |
| Professional work | ||
| Finance Agent (v2) | 58.6% | 56.3% |
| EMB | 75.9% | 65.5% |
| HLAB-Vals-1.1 | 9.2% | 13.3% |
| Legal Research Bench | 50.5% | 38.0% |
| MedCode | 49.8% | 41.1% |
| MedScribe | 91.4% | 85.3% |
| Public Benefits Bench v1.1 | 70.6% | 67.6% |
| SAGE | 91.3% | 83.3% |
| Tax Agent Bench | 44.5% | 28.5% |
| Novel problems | ||
| MysteryMechanism | 49.5% | 21.6% |
Tests at least two of these models have taken. Bold rows are the core tests.