Kimi K2.6vsKimi K2 Thinking
Across 6 shared benchmarks, Kimi K2.6 leads overall: Kimi K2.6 wins 5, Kimi K2 Thinking wins 1, with 0 ties and an average score difference of +7.63.
Kimi K2.6
Moonshot AI · 2026-04-20 · Reasoning model
Kimi K2 Thinking
Moonshot AI · 2025-11-06 · Reasoning model
Kimi K2.65 wins(83%)(17%)1 winKimi K2 Thinking
Benchmark scores
Grouped by capability, sorted by largest gap within each. 6 shared benchmarks.
Coding and Software Engineer
Kimi K2.6 2/2| Benchmark | Kimi K2.6 | Kimi K2 Thinking | Diff |
|---|---|---|---|
| SWE-bench Verified | 80.2013 / 108Thinking (With Tools) | 71.3051 / 108 | +8.90 |
| LiveCodeBench | 89.607 / 120Thinking (No Tools) | 83.1022 / 120 | +6.50 |
General Knowledge
Kimi K2.6 2/2| Benchmark | Kimi K2.6 | Kimi K2 Thinking | Diff |
|---|---|---|---|
| GPQA Diamond | 90.5016 / 178Thinking (No Tools) | 84.5052 / 178 | +6 |
| HLE | 549 / 157Thinking (With Tools + Internet) | 5116 / 157 | +3 |
AI Agent - Information Search
Kimi K2.6 1/1| Benchmark | Kimi K2.6 | Kimi K2 Thinking | Diff |
|---|---|---|---|
| BrowseComp | 83.2010 / 45Thinking (With Tools + Internet) | 60.2030 / 45 | +23 |
Claw-style Agent Evaluation
Kimi K2 Thinking 1/1| Benchmark | Kimi K2.6 | Kimi K2 Thinking | Diff |
|---|---|---|---|
| Claw Bench | 80.9019 / 29Thinking (With Tools) | 82.5017 / 29Thinking (With Tools) | -1.60 |
Specs
| Field | Kimi K2.6 | Kimi K2 Thinking |
|---|---|---|
| Publisher | Moonshot AI | Moonshot AI |
| Release date | 2026-04-20 | 2025-11-06 |
| Model type | Reasoning model | Reasoning model |
| Architecture | MoE | MoE |
| Parameters | 1T | 1T |
| Context length | 256K | 256K |
| Max output | Not available | Not available |
API pricing
Prices use DataLearner records when available; missing fields are not inferred.
| Item | Kimi K2.6 | Kimi K2 Thinking |
|---|---|---|
| Text input | $0.95 / 1M tokens | Not public |
| Text output | $4 / 1M tokens | Not public |
| Cache read | $0.16 / 1M tokens | Not public |
| Cache write | $0.95 / 1M tokens | Not public |
One or both models have incomplete public pricing.
Summary
- Kimi K2.6leads in:Coding and Software Engineer (2/2), General Knowledge (2/2), AI Agent - Information Search (1/1)
- Kimi K2 Thinkingleads in:Claw-style Agent Evaluation (1/1)
On average across the 6 shared benchmarks, Kimi K2.6 scores 7.63 higher.
Largest single-benchmark gap: BrowseComp — Kimi K2.6 83.20 vs Kimi K2 Thinking 60.20 (+23).
Page generated from structured model, pricing and benchmark records. No real-time LLM is used to write the prose.