All comparisons
Google-Proof Q&A — 87.9 MMLU PRO — 87.1 Massive Multi-discipline Multimodal Understanding — 84.3 Instruction-Following Evaluation — 93.9 BFCL — 68.3
Kimi K2.6 vs Kimi K2.5
Side-by-side comparison of specifications, capabilities, and pricing.
| Released | 2026-04-13 | 2026-03-15 |
| Context window | 262K | 256K |
| Parameters | 1T (MoE, 32B active) | 1T (MoE, 384 experts) |
| Architecture | Mixture of Experts (MoE) | mixture of experts |
| License | Proprietary | MIT |
| Knowledge cutoff | — | — |
Capabilities | ||
| Vision | ||
| Multimodal | ||
| Reasoning | ||
| Function calling | ||
| Tool use | ||
| Structured Outputs | ||
| Code execution | ||
Availability | ||
| Providers | ||
Benchmarks