Your first uncensored DeepSeek V4.1 Flash request
DeepSeek V4.1 Flash Outlaw answers on an OpenAI-compatible endpoint and an Anthropic-compatible one. Any client that takes a base URL works: change the URL, the key and the model id.
The three values
| Setting | Value |
|---|---|
| Base URL (OpenAI format) | https://wildwestapi.com/v1 |
| Base URL (Anthropic format) | https://wildwestapi.com |
| Model id | deepseek-v4.1-flash-outlaw |
| API key | sk-ww-... from your dashboard |
curl
curl https://wildwestapi.com/v1/chat/completions \ -H "Authorization: Bearer sk-ww-..." \ -H "Content-Type: application/json" \ -d '{ "model": "deepseek-v4.1-flash-outlaw", "messages": [{"role": "user", "content": "Hello"}] }'
Python
# pip install openai from openai import OpenAI client = OpenAI( base_url="https://wildwestapi.com/v1", api_key="sk-ww-...", ) resp = client.chat.completions.create( model="deepseek-v4.1-flash-outlaw", messages=[{"role": "user", "content": "Hello"}], ) print(resp.choices[0].message.content)
JavaScript
// npm install openai import OpenAI from "openai"; const client = new OpenAI({ baseURL: "https://wildwestapi.com/v1", apiKey: "sk-ww-...", }); const resp = await client.chat.completions.create({ model: "deepseek-v4.1-flash-outlaw", messages: [{ role: "user", content: "Hello" }], }); console.log(resp.choices[0].message.content);
A tool call
This is the request the model passed before it went on sale, on 18 October 2026:
curl https://wildwestapi.com/v1/chat/completions \ -H "Authorization: Bearer sk-ww-..." \ -H "Content-Type: application/json" \ -d '{ "model": "deepseek-v4.1-flash-outlaw", "messages": [{"role": "user", "content": "What is the weather in Auckland? Use the tool."}], "tools": [{"type": "function", "function": { "name": "get_weather", "description": "Current weather for a city", "parameters": {"type": "object", "properties": {"city": {"type": "string"}}, "required": ["city"]} }}] }'
It came back with finish_reason: "tool_calls" and a call to get_weather with the arguments {"city": "Auckland"}. Send the tool's result back as a tool message and the model answers from it, the same loop every agent runs.
Limits worth knowing
| DeepSeek V4.1 Flash Outlaw | |
|---|---|
| Context | 1,048,576 tokens |
| Longest reply | 393,216 tokens |
| Tool calling | Yes |
| Reasoning | On by default; set the effort |
| Images | No, text only |
It is text only, so for screenshots use Hipshooter, which reads them on the same key. Current prices are on the model page and pricing.
Claude Code and other agents
Claude Code uses the Anthropic format: set ANTHROPIC_BASE_URL to https://wildwestapi.com and ANTHROPIC_MODEL to deepseek-v4.1-flash-outlaw, as in the Claude Code guide. Codex CLI uses the Responses format; see Codex CLI on DeepSeek V4.1 Flash.
FAQ
What is the model id for uncensored DeepSeek V4.1 Flash?
deepseek-v4.1-flash-outlaw, on https://wildwestapi.com/v1 for OpenAI clients or https://wildwestapi.com for Anthropic clients.
Does it support function calling?
Yes. It returned a correct get_weather call before it went on sale, and it takes the standard OpenAI tools array.
Does it stream?
Yes. Set "stream": true as with any OpenAI-compatible endpoint.
Can I send it images?
No. It is text only. Hipshooter and High Noon take images on the same key.