Up to four models
Kimi K2.6 vs Grok 4.3
On the index (GPT-4 = 100), Kimi K2.6 scores 152 and Grok 4.3 135. Side by side on the same third-party tests, with price and context. The link in your address bar shares this exact comparison.
Kimi K2.6
152136–169
WeightsOpen
ReleasedApr 2026
Price in / out$0.65 / $3.41
Context262K
Core tests taken4 of 9
Grok 4.3
135115–156
WeightsClosed
ReleasedApr 2026
Price in / out$1.25 / $2.5
Context1M
Core tests taken1 of 9
Kimi K2.6 leads Grok 4.3 by 17 points, ahead on 1 of 1 shared core tests, at a similar price.
Test by testthird-party results
Kimi K2.6Grok 4.3
Every result as a table
| Test | Kimi K2.6 | Grok 4.3 |
|---|---|---|
| Reasoning & knowledge | ||
| LiveBench Data Analysis | 65.1% | 55.8% |
| LiveBench IF | 64.4% | 62.7% |
| LiveBench Language | 75.1% | 73.6% |
| LiveBench Reasoning | 79.4% | 70.8% |
| GPQA Diamond (Vals AI) | 89.1% | 91.4% |
| MMLU Pro | 87.6% | 85.8% |
| MMMU Pro | 86.3% | 83.1% |
| Maths | ||
| LiveBench Mathematics | 84.3% | 84.3% |
| Coding | ||
| LiveBench Agentic Coding | 46.9% | 18.5% |
| LiveBench Coding | 78.6% | 69.9% |
| Code Migration | 27.8% | 6.8% |
| LiveCodeBench | 86.8% | 84.5% |
| ProgramBench | 0.0% | 0.0% |
| SWE-bench | 76.2% | 71.4% |
| Vibe Code Bench v1.1 | 37.9% | 19.4% |
| Professional work | ||
| Finance Agent (v2) | 44.9% | 37.7% |
| CorpFin v2 | 66.7% | 68.5% |
| EMB | 57.9% | 18.4% |
| LegalBench | 84.7% | 84.5% |
| Legal Research Bench | 15.9% | 15.4% |
| MedCode | 40.1% | 38.1% |
| MedScribe | 78.1% | 74.4% |
| MortgageTax | 65.8% | 48.3% |
| Public Benefits Bench v1.1 | 56.6% | 51.7% |
| Tax Agent Bench | 12.4% | 6.2% |
| TaxEval v2 | 74.7% | 70.8% |
Tests at least two of these models have taken. Bold rows are the core tests.