- First ~3T scale open model
- Most capable open model ever
- It's Fable/Sol class open model beating Opus 4.8
- 1M Context · In $3/M Out $15/M Cache read $0.3/M
What an exciting time to be alive!! Try now!
Kimi K3 at a glance
Long-horizon coding & knowledge work with 1M context.
| Spec | Kimi K3 |
|---|---|
| Model ID | moonshotai/kimi-k3 |
| Context window | 1M tokens |
| Released | July 16, 2026 |
| Intelligence Index | 43.6 |
| Coding Index | 76.2 |
| Input | $3 per 1M tokens |
| Output | $15 per 1M tokens |
| Cache read | $0.30 per 1M tokens |
| Effective input in an agent loop | $1.11 per 1M tokens |
| Plans | Go and above |
Specs, pricing and plans are from the Kimi K3 model page as of September 28, 2026, including any deal running that day, so they can differ from the launch pricing above. Intelligence and Coding Index scores are from Artificial Analysis.
Use Kimi K3
1cmd --model moonshotai/kimi-k3Or type /model in a session and pick it. You can switch mid-session without losing context.
What real work costs on Kimi K3
| Task | Tokens (in · out) | Cost |
|---|---|---|
| Quick lookup / one-liner | 8K · 1K | $0.02 |
| Review a 500-line PR | 60K · 4K | $0.13 |
| Fix a bug (agent loop) | 180K · 12K | $0.36 |
| Refactor a module | 320K · 20K | $0.64 |
| Full-repo agent run | 900K · 45K | $1.48 |
Try Command Code
The best coding agent for open models, in your terminal or on your desktop.
1npm i -g command-codeDesktop App · Docs · X · Discord
