Kimi K3.1 Uncensored is live
Kimi K3.1 Uncensored is now our flagship. It is Moonshot AI's open-weight model with its refusal behavior removed, built for long agent runs, research and very long outputs.
What it is good at
Kimi K3.1 holds a lot in view and writes a lot in one go. It reads up to 1M tokens, so a coding agent can keep most of a repository in context, and it can write up to 944K tokens in a single reply, which is enough for whole reports, long documents or large refactors without stitching.
Specs and pricing
| Kimi K3.1 | |
|---|---|
| Model id | kimi-k3.1-uncensored |
| Context | 1,048,576 tokens |
| Max output | 943,718 tokens |
| Input | $2.00 per 1M tokens |
| Cached input | $0.20 per 1M tokens |
| Output | $10.00 per 1M tokens |
| Reasoning effort | Not adjustable |
Coming from Kimi K3.0
kimi-k3.1-uncensored replaces kimi-k3.0-uncensored, which now returns model_not_found. Change the model id and nothing else.
Get started
Create an API key, add credits from $5, and send a request to kimi-k3.1-uncensored at https://api.silkcompute.com. The model page has a prompt you can paste into your coding agent to set it up for you, plus code for Python, JavaScript, cURL, Claude Code, Codex and OpenCode.
