Up to four models
GPT-5.5 vs Kimi K2.6
On the index (GPT-4 = 100), GPT-5.5 scores 172 and Kimi K2.6 152. Side by side on the same third-party tests, with price and context. The link in your address bar shares this exact comparison.
GPT-5.5
172156–188
WeightsClosed
ReleasedApr 2026
Price in / out$5.0 / $30
Context1.05M
Core tests taken6 of 9
Kimi K2.6
152136–169
WeightsOpen
ReleasedApr 2026
Price in / out$0.65 / $3.41
Context262K
Core tests taken4 of 9
GPT-5.5 leads Kimi K2.6 by 20 points, ahead on 4 of 4 shared core tests, at 8.4× the price.
Test by testthird-party results
GPT-5.5Kimi K2.6
Every result as a table
| Test | GPT-5.5 | Kimi K2.6 |
|---|---|---|
| Reasoning & knowledge | ||
| GPQA Diamond (Epoch AI) | 94.0% | 90.8% |
| LiveBench Data Analysis | 81.6% | 65.1% |
| LiveBench IF | 70.7% | 64.4% |
| LiveBench Language | 87.4% | 75.1% |
| LiveBench Reasoning | 89.6% | 79.4% |
| GPQA Diamond (Vals AI) | 93.2% | 89.1% |
| MMLU Pro | 88.1% | 87.6% |
| MMMU Pro | 88.3% | 86.3% |
| EBR-bench | 34.3% | 2.4% |
| SimpleQA Verified | 63.0% | 34.9% |
| Maths | ||
| FrontierMath (tiers 1-3) | 85.3% | 57.2% |
| AIME-style maths (OTIS mock) | 100.0% | 96.1% |
| LiveBench Mathematics | 95.9% | 84.3% |
| Chess puzzles | 63.0% | 20.0% |
| FrontierMath (2025 set) | 51.7% | 39.0% |
| FrontierMath tier 4 (2025 set) | 35.4% | 14.6% |
| FrontierMath (tier 4) | 72.5% | 25.6% |
| Coding | ||
| SWE-bench Verified | 80.6% | 76.6% |
| LiveBench Agentic Coding | 54.0% | 46.9% |
| LiveBench Coding | 82.2% | 78.6% |
| Code Migration | 45.2% | 27.8% |
| LiveCodeBench | 85.3% | 86.8% |
| ProgramBench | 0.5% | 0.0% |
| SWE-bench | 82.6% | 76.2% |
| Vibe Code Bench v1.1 | 69.8% | 37.9% |
| WeirdML | 84.9% | 55.9% |
| Agents | ||
| DTBench | 96.0% | 90.9% |
| Furniture assembly | 44.2% | 21.7% |
| LMCA | 54.3% | 37.3% |
| Mystery game puzzles | 56.0% | 18.0% |
| Professional work | ||
| Finance Agent (v2) | 51.8% | 44.9% |
| CorpFin v2 | 68.4% | 66.7% |
| EMB | 64.5% | 57.9% |
| LegalBench | 86.5% | 84.7% |
| Legal Research Bench | 40.4% | 15.9% |
| MedCode | 49.1% | 40.1% |
| MedScribe | 86.9% | 78.1% |
| MortgageTax | 68.8% | 65.8% |
| Public Benefits Bench v1.1 | 60.9% | 56.6% |
| Tax Agent Bench | 21.8% | 12.4% |
| TaxEval v2 | 75.0% | 74.7% |
Tests at least two of these models have taken. Bold rows are the core tests.