Kimi K3vsKimi K2 Thinking
Across 5 shared benchmarks, Kimi K3 leads overall: Kimi K3 wins 5, Kimi K2 Thinking wins 0, with 0 ties and an average score difference of +101.82.
Kimi K3
Moonshot AI · 2026-07-16 · Reasoning model
Kimi K2 Thinking
Moonshot AI · 2025-11-06 · Reasoning model
Kimi K35 wins(100%)(0%)0 winsKimi K2 Thinking
Benchmark scores
Grouped by capability, sorted by largest gap within each. 5 shared benchmarks.
AI Agent - Information Search
Kimi K3 1/1| Benchmark | Kimi K3 | Kimi K2 Thinking | Diff |
|---|---|---|---|
| BrowseComp | 91.201 / 55Max (With Tools + Internet) | 60.2037 / 55Thinking (With Tools) | +31 |
Commonsense Reasoning
Kimi K3 1/1| Benchmark | Kimi K3 | Kimi K2 Thinking | Diff |
|---|---|---|---|
| SimpleBench | 60.7029 / 92最高（无工具） | 39.6045 / 67Normal (No Tools) | +21.10 |
General Evaluation
Kimi K3 1/1| Benchmark | Kimi K3 | Kimi K2 Thinking | Diff |
|---|---|---|---|
| GPQA Diamond | 93.5015 / 270最高（无工具） | 84.50106 / 270Thinking (No Tools) | +9 |
General Knowledge
Kimi K3 1/1| Benchmark | Kimi K3 | Kimi K2 Thinking | Diff |
|---|---|---|---|
| HLE | 5616 / 189Max (With Tools) | 5129 / 189Parallel Thinking (With Tools) | +5 |
Writing and Creative Capabilities
Kimi K3 1/1| Benchmark | Kimi K3 | Kimi K2 Thinking | Diff |
|---|---|---|---|
| Creative Writing | 2,0712 / 99Normal (No Tools) | 1,62831 / 99Normal (No Tools) | +443 |
Specs
| Field | Kimi K3 | Kimi K2 Thinking |
|---|---|---|
| Publisher | Moonshot AI | Moonshot AI |
| Release date | 2026-07-16 | 2025-11-06 |
| Model type | Reasoning model | Reasoning model |
| Architecture | MoE | MoE |
| Parameters | 2.8T | 1T |
| Context length | 1M | 256K |
| Max output | 1M | Not available |
API pricing
Prices use DataLearner records when available; missing fields are not inferred.
| Item | Kimi K3 | Kimi K2 Thinking |
|---|---|---|
| Text input | ¥20 / 1M tokens | $0.6 / 1M tokens |
| Text output | ¥100 / 1M tokens | $2.5 / 1M tokens |
| Cache read | ¥2 / 1M tokens | $0.3 / 1M tokens |
| Cache write | Not public | $0.6 / 1M tokens |
Summary
- Kimi K3leads in:AI Agent - Information Search (1/1), Commonsense Reasoning (1/1), General Evaluation (1/1), General Knowledge (1/1), Writing and Creative Capabilities (1/1)
On average across the 5 shared benchmarks, Kimi K3 scores 101.82 higher.
Largest single-benchmark gap: Creative Writing — Kimi K3 2,071 vs Kimi K2 Thinking 1,628 (+443).
Page generated from structured model, pricing and benchmark records. No real-time LLM is used to write the prose.