Run Kimi K3.1 Pro Outlaw in Zed
Run kimi-k3.1-pro-outlaw in Zed through Wild West: one key, a hard spend cap, and a model that answers instead of refusing.
| Model id | kimi-k3.1-pro-outlaw |
|---|---|
| Context window | 1,048,576 tokens (1M) |
| Tool calling | Supported |
| Price per 1M tokens | $7.50 input, $0.80 on a cache hit, $37.50 output |
| Intelligence | 55 on the Artificial Analysis Intelligence Index v4.3.2, for the base model |
| Web search | Add :online to the id |
Set it up
- Declare an openai_compatible provider in settings.json with kimi-k3.1-pro-outlaw in available_models.
- Add the key through Agent Settings, Add Provider, so it lives in the OS keychain rather than the file.
- Pick the model in the Agent Panel.
{
"language_models": {
"openai_compatible": {
"wildwest": {
"api_url": "https://wildwestapi.com/v1",
"available_models": [
{
"name": "kimi-k3.1-pro-outlaw",
"display_name": "Kimi K3.1 Pro Outlaw (Wild West)",
"max_tokens": 1048576,
"max_output_tokens": 32768,
"capabilities": { "tools": true }
}
]
}
}
}
}
Kimi K3.1 Pro Outlaw has its own page with every limit and price.
Uncensored Kimi K3.1 Pro sets it against the rest of the line, and uncensored Kimi goes through the Kimi releases one by one.
What a session costs
For a 25-turn agent task (about 750K input tokens re-read across turns, 85% of them cache hits, 25K output), Kimi K3.1 Pro Outlaw comes to about $2.29 at today's prices; the cache price does most of the work, since re-read context is billed at $0.80 rather than $7.50 per million.
FAQ
Can Kimi K3.1 Pro Outlaw edit files and run commands in Zed?
Yes. Zed acts through tool calls and this model supports tool calling, so it can read, edit and run, not just chat.
What does a session cost?
About $2.29 for a 25-turn agent task (about 750K input tokens re-read across turns, 85% of them cache hits, 25K output), at $7.50 in, $37.50 out and $0.80 per million on cache hits. Put a spend cap on the key so a long run cannot overrun it.
Is it uncensored?
Kimi K3.1 Pro, uncensored: the strongest model on the line, with a million tokens of context, tool calling, reasoning and vision.