Connect SillyTavern to Outrun.
Connect SillyTavern to Outrun with one key, then test the connection.
Before you start
Have your API key and some Outrun credit ready. Keep your key private.
Optional setup helperUse an AI assistant to walk through these steps
Copy the setup instructions into an AI assistant you already use. Review any file changes it suggests before allowing them.
Enter your API key directly in the app or local settings. Keep it out of the assistant’s chat.
Connection settingsShow
API type: Chat Completion
Chat Completion Source: Custom (OpenAI-compatible)
Custom endpoint: https://api.outrun.to/v1
First model ID: glm-5.3-flash-uncensoredGet your key from Outrun → API keys and enter it privately in the app or local settings. Test with “Reply with OK only.” Tests use credit; check Activity before retrying.
Setup Steps
- 1Get your Outrun key
Open API keys on the Outrun website, or open @outrunAPI_Bot and choose My API key. Paste your key only into SillyTavern’s private API-key field.
- 2Open API Connections
In SillyTavern, open API Connections. Choose Chat Completion as the API type.
- 3Choose the Custom source
For Chat Completion Source, choose Custom (OpenAI-compatible). This guide uses the chat connection, not a text-completion connection.
- 4Enter the endpoint and key
Enter https://api.outrun.to/v1 as the custom endpoint. Enter your Outrun key separately in the key field. Do not append /chat/completions: SillyTavern adds the operation path.
- 5Connect and choose a model
Click Connect. If the model dropdown loads, choose an available model. Outrun handles authenticated GET /v1/models; the IDs available to your key may vary. If no list is returned, type an exact model ID from the recommendations below.
- 6Send one short test
Start a chat and send: Reply with OK only. This uses Outrun credit. If a reply is empty or the request is interrupted, check Activity before trying another route or model.
General chat models
Choose a model you already use. Open its detail page for current prices, limits and route options.
| Model | Model ID | Details |
|---|---|---|
| Claude Sonnet 5.5 | claude-sonnet-5-5 | Model details |
| GPT-6.1 Sol | gpt-6.1-sol | Model details |
| DeepSeek V4 Pro | deepseek-v4-pro | Model details |
Uncensored and abliterated text models
Choose an uncensored or abliterated text model. Open its detail page to compare routes, limits and current prices.
| Model | Model ID | Details |
|---|---|---|
| GLM-5.3 Flash Uncensored | glm-5.3-flash-uncensored | Model details |
| glm-5.3-abliterated | glm-5.3-abliterated | Model details |
| DeepSeek V4 Pro Uncensored | deepseek-v4-pro-uncensored | Model details |
| MiMo V2.6 Flash Uncensored | mimo-v2.6-flash-uncensored | Model details |
Context, response length and streaming
Context (tokens) is the total conversation budget. Keep it within the selected model’s context limit, including the room reserved for its response. SillyTavern subtracts Response (tokens) from that budget for the prompt.
Keep Response (tokens) at its default or a smaller value within the model’s output limit. Enable Streaming unless the selected model’s setup data says it is off. Leave temperature and other unspecified controls at their defaults.
The limits below come from the same setup data as the model pages. Route IDs can have different limits or capabilities: switching routes means checking that route’s row again. Larger prompts can fall into another pricing tier; check the linked model page.
| Model / route ID | Context limit | Response limit | Streaming |
|---|---|---|---|
| claude-sonnet-5-5 | 1,000,000 tokens | 128,000 tokens | On |
| gpt-6.1-sol | 1,050,000 tokens | 128,000 tokens | On |
| deepseek-v4-pro | 1,000,000 tokens | 393,216 tokens | On |
| glm-5.3-flash-uncensored | 1,000,000 tokens | 65,536 tokens | On |
| glm-5.3-abliterated | 1,000,000 tokens | 130,000 tokens | On |
| deepseek-v4-pro-uncensored | 1,000,000 tokens | 65,536 tokens | On |
| mimo-v2.6-flash-uncensored | 1,048,576 tokens | 65,536 tokens | On |
| glm-5.3-flash-uncensored-2 | 1,000,000 tokens | 65,536 tokens | On |
| glm-5.3-flash-uncensored-3 | 1,000,000 tokens | 32,768 tokens | On |
Optional: download the connection reference
Download a small JSON reference with the endpoint, source and first model ID. It contains no API key. This is a copy-paste reference, not an importable SillyTavern profile. Copy its values into API Connections, enter your key separately, then click Connect. You can save the configured connection inside SillyTavern using its Connection Profiles feature.
{
"api": "Chat Completion",
"source": "Custom (OpenAI-compatible)",
"endpoint": "https://api.outrun.to/v1",
"model": "glm-5.3-flash-uncensored"
}Test your key first
This optional cURL request separates a key or balance problem from an app setting. Put your key in OUTRUN_API_KEY privately before running it. One model request uses credit; check Activity before retrying.
curl https://api.outrun.to/v1/chat/completions \
-H "Authorization: Bearer $OUTRUN_API_KEY" \
-H "Content-Type: application/json" \
-d '{"model":"glm-5.3-flash-uncensored","messages":[{"role":"user","content":"Reply with OK only."}],"max_tokens":16}'If something goes wrong
Key missing or rejected (401)
Copy your current key from API keys. Paste it only into the app’s key field or your local key variable, then save and reopen the app.
Not enough available credit (402)
Check Balance and the key’s spending cap. Another request may have temporarily reserved credit; read the message before adding funds.
Model not found (400/404)
Copy a current model ID from Models & Pricing. A model name or a route label is not an API model ID.
Empty replies or an interrupted request
Check Activity first. Confirm the response-length setting is within the model’s output limit. If needed, choose another listed route or model; do not repeatedly submit the same request.
Busy or rate limited (429)
Wait for the active request to finish or the rate limit to reset. Avoid repeatedly submitting the same request.
Model choices and service notes
Use a currently available chat model from Models & Pricing. Connection tests and chats use prepaid Outrun credit. Check Activity before repeating an interrupted request.