Kimi K3 vs Grok 4.5
Grok 4.5 is 1.1× cheaper for typical traffic ; Kimi K3 scores higher on the Artificial Analysis intelligence index (43.6 vs 38.8).
Prices checked Oct 21, 2026
| Kimi K3 | Grok 4.5 | |
|---|---|---|
| Developer | Moonshot AI | xAI |
| Input / 1M tokens | $0.239 | $2.00 |
| Output / 1M tokens | $11.94 | $6.00 |
| Cached input / 1M | $0.21 | $0.30 |
| Context window | 1.05M | 500K |
| Max output | 944K | 450K |
| Intelligence index | 43.6 | 38.8 |
| Image input | Yes | Yes |
| Tool calling | Yes | Yes |
| Reasoning mode | Yes | Yes |
| Open weights | Yes | No |
Monthly cost for common workloads
| Workload | Kimi K3 | Grok 4.5 |
|---|---|---|
| Customer chatbot 1,000 conversations a day, 8 turns each | $1,082 | $857 |
| Document Q&A (RAG) 500 questions a day over retrieved documents | $111 | $210 |
| Coding / tool-using agent 200 tasks a day, long tool-heavy contexts | $365 | $405 |
| Content generation 30 long articles a day | $27.50 | $18.90 |
| Classification / extraction 100,000 short items a day | $752 | $2,070 |
Estimates include cached-input discounts where a model offers them. Try your own numbers in the cost calculator.
Which one to choose
- Lowest cost: Grok 4.5 — 1.1× cheaper on a 3:1 input/output mix.
- Longer documents: Kimi K3 reads up to 1.05M tokens at once.
- Capabilities: Kimi K3 — vision, tools, reasoning, JSON, open weights; Grok 4.5 — vision, tools, reasoning, JSON.