Kimi K3.1 Pro
Moonshot AI
Moonshot AI's latest open-weights model, built off of Kimi K3.
Token pricing and context
- Input
- Not reported
- Cached input
- Not reported
- Output
- Not reported
- Context window
- 1000000
Benchmark results
MM1.0
Overall score: 89 · Scale 0–100 · Higher is better
- Reasoning: 100
- Coding: 92
- Chat: 90
- Long Context: 95
- Vibe: 85
- Accuracy: 100
- General Knowledge: 37.5
Evaluated: 2026-10-24
Hallucimeter 1.0
Overall score: 7 · Scale 0–100 · Lower is better
- Hallucinations: 7
Evaluated: 2026-10-24
Updated 2026-10-24. Read benchmark methodology.