---
title: "SillyTavern Custom OpenAI-compatible endpoint setup | BroadRouter"
description: "Connect SillyTavern to BroadNet Inference with the Custom (OpenAI-compatible) Chat Completion source, the right model IDs, prompt post-processing and token limits."
url: https://broadrouter.com/guides/sillytavern-custom-openai-endpoint/
lang: en
updated: 2026-11-16
publisher: BroadRouter (a BroadNet site); API by BroadNet Inference, https://inference.broadnet.ai
---

# Connect SillyTavern to BroadNet Inference (Custom OpenAI-compatible)

Updated 2026-11-16

In SillyTavern's API Connections panel, choose Chat Completion, then Custom (OpenAI-compatible) as the Chat Completion Source. Set Custom Endpoint to https://inference.broadnet.ai/v1, paste your BroadNet key, type a model ID such as broadnet/qwen3.8-27b in Enter a Model ID, click Connect, then Test Message. Set Max Response Length to 1,000 tokens or more.

SillyTavern is an open-source chat front end, popular for character chat and roleplay. Its **Custom (OpenAI-compatible)** source connects the Chat Completion API type to any server with the generic OpenAI schema. BroadNet Inference serves standard chat completions at `https://inference.broadnet.ai/v1`. BroadRouter is BroadNet’s own site; the steps follow [SillyTavern’s docs](https://docs.sillytavern.app/usage/api-connections/openai/) and its 1.19.0 release.

## Setup

1.  **Get a key.** Create a pay-as-you-go key in the BroadNet console, buy an hourly key, or buy an [unlimited plan](https://broadrouter.com/guides/unlimited-llm-api-sessions/).
    
2.  Open the **API Connections** panel.
    
3.  Set **API** to **Chat Completion**.
    
4.  Set **Chat Completion Source** to **Custom (OpenAI-compatible)**.
    
5.  Fill in the custom form:
    
    | Field | Value |
    | --- | --- |
    | Custom Endpoint (Base URL) | `https://inference.broadnet.ai/v1` |
    | Custom API Key | your BroadNet key |
    | Enter a Model ID | `broadnet/qwen3.8-27b` |
    
6.  Click **Connect**.
    
7.  **Test it.** Click **Test Message**. SillyTavern sends a simple prompt to the model; a reply means the connection works.
    

Don’t add `/chat/completions` to the URL. SillyTavern’s form says that suffix is added automatically.

Type the model ID rather than picking from **Available Models**. That list comes from `/v1/models`, which BroadNet’s docs describe as catalogue metadata for monitors. When we checked it on 15 November 2026 it listed only the two transcription models, which can’t chat.

## Which model to pick

| Model | ID | Context | Notes |
| --- | --- | --- | --- |
| DeepSeek V4.1 Flash | `broadnet/deepseek-v4.1-flash` | 1,048,576 | Long chats; the model on the unlimited plans |
| Qwen3.8 27B | `broadnet/qwen3.8-27b` | 262,144 | Images; tested in Arabic by BroadNet |
| GLM 5.3 | `broadnet/glm-5.3` | 1,048,576 | Text only; tools |
| BroadNet Gemma 4 VL 26B | `broadnet/gemma-4-26b-a4b` | 32,768 | Runs on the $5 free credit and the hourly key |

An unlimited plan suits long sessions because you don’t pay per token: a Meteor hour costs $1, an Orbit week $15 and a Full Moon month $60. Each plan session runs one request at a time.

## Settings worth changing

-   **Max Response Length (tokens).** DeepSeek V4.1 Flash, GLM 5.3 and Qwen3.8 27B reason before they answer, and the reasoning counts toward this limit. Set 1,000 or more, or you may get blank replies.
-   **Context Size (tokens).** Match it to the model in the table. SillyTavern can’t read it from BroadNet.
-   **Prompt Post-Processing.** If a model rejects a prompt because of several system messages or role order, try **Merge consecutive roles** first, then **Semi-strict**. SillyTavern’s docs note that these options drop tool calls unless you pick the “with tools” variant.
-   **Bypass status check.** If SillyTavern warns that the API isn’t working but Test Message succeeds, tick this.

## Troubleshooting

-   **401.** The key is wrong, expired or out of allowance. Paste it again and click Connect.
-   **404 or “model not found”.** The URL must end in `/v1`, and the ID must include the `broadnet/` prefix. Paid models need a top-up; the $5 credit covers only `broadnet/gemma-4-26b-a4b` and the two Transcribe models.
-   **Blank replies.** Raise the response length. [Why answers come back empty](https://broadrouter.com/guides/reasoning-tokens-empty-response/).
-   **429.** A pay-as-you-go key allows 30 requests a minute and 4 in flight, and the hourly key and plan sessions run one stream. See [errors and retries](https://broadrouter.com/guides/errors-and-retries/).

## Terms in this guide

[Pay-as-you-go](https://broadrouter.com/glossary/pay-as-you-go/)

Pay-as-you-go is a pricing model in which you pay only for what you use, metered per unit such as tokens, requests or audio minutes, with no subscription or minimum commitment. Most LLM APIs bill this way, from a prepaid balance or a periodic invoice.

[Token](https://broadrouter.com/glossary/token/)

A token is the unit of text a language model reads and writes: a word, part of a word, a punctuation mark or a byte sequence produced by the model's tokenizer. LLM APIs count, limit and usually bill usage in tokens.

[Tool calling](https://broadrouter.com/glossary/tool-calling/)

Tool calling, also called function calling, is an LLM API feature in which the model returns a structured request to run a function the developer has defined, with JSON arguments. The application runs the function and sends the result back; the model never runs code itself.

[API key](https://broadrouter.com/glossary/api-key/)

An API key is a secret string that identifies and authorizes a client calling an API. LLM APIs usually expect it in an Authorization: Bearer header, and anyone who holds the key can make requests that are billed to the account that owns it.

[Base URL](https://broadrouter.com/glossary/base-url/)

A base URL is the fixed root address of an API, such as https://api.example.com/v1, that a client puts in front of every endpoint path, such as /chat/completions. In OpenAI-compatible SDKs, changing the base URL, together with the API key and model ID, points the same code at a different provider.

[Chat Completions API](https://broadrouter.com/glossary/chat-completions-api/)

The Chat Completions API is OpenAI's HTTP endpoint, POST /v1/chat/completions, that takes a model name and a list of role-tagged messages and returns the model's next message. Its request and response shape became the common format that OpenAI-compatible providers and SDKs reuse.

[Concurrency](https://broadrouter.com/glossary/concurrency/)

Concurrency is the number of requests a client has in flight at the same time. LLM APIs often cap it per key or plan, separately from requests per minute, because each open request holds GPU capacity until its response has finished.

[Context window](https://broadrouter.com/glossary/context-window/)

A context window is the maximum number of tokens a language model can work with in a single request. It covers the prompt, conversation history, tool definitions, images and the generated output, including any reasoning tokens. Anything that does not fit must be cut, summarised or split across several requests.

## Questions

### Why is the Available Models list wrong or empty for BroadNet?

SillyTavern fills it from /v1/models. When we checked on 15 November 2026, BroadNet's public /v1/models listed only its two transcription models. Type the chat model ID in Enter a Model ID instead.

### Which BroadNet model is best for long roleplay chats?

DeepSeek V4.1 Flash has a 1,048,576-token context and is the model on BroadNet's flat-rate unlimited plans. Qwen3.8 27B is the one BroadNet tested in Arabic, including Gulf and Egyptian dialects.

### Why do replies come back blank in SillyTavern?

BroadNet's chat models except Gemma 4 VL 26B reason before answering, and the reasoning counts toward max\_tokens. SillyTavern's Max Response Length sets that limit, so raise it to 1,000 tokens or more.

### Are there content rules?

Yes. BroadNet's Acceptable Use Policy applies to everything sent through the API, and it explicitly covers fictional and synthetic content in its ban on sexual content involving minors.

## Related guides

-   [Use Jan with BroadNet Inference as a custom OpenAI-compatible provider](https://broadrouter.com/guides/jan-custom-endpoint/)
-   [Connect Open WebUI to BroadNet Inference](https://broadrouter.com/guides/open-webui-openai-api/)
-   [Why a reasoning model returns an empty answer](https://broadrouter.com/guides/reasoning-tokens-empty-response/)
-   [BroadNet Inference error codes and when to retry](https://broadrouter.com/guides/errors-and-retries/)
-   [All tools that work with BroadNet Inference](https://broadrouter.com/integrations/): SDKs, coding agents, chat UIs and automation tools, with exact settings for each

Try it on your real workload. New accounts start with $5 of credit for Gemma and transcription. DeepSeek, GLM and Qwen3.8 need a top-up.

[Get started](https://inference.broadnet.ai/account/?utm_source=broadrouter&utm_medium=referral&utm_campaign=guides-sillytavern-custom-openai-endpoint&utm_content=page-cta) ·

**Sources**

-   [SillyTavern docs, Chat Completions API connections, Custom OpenAI-compatible endpoint (checked 2026-11-15)](https://docs.sillytavern.app/usage/api-connections/openai/)
-   [SillyTavern source, Custom endpoint form in public/index.html, release branch (checked 2026-11-15)](https://github.com/SillyTavern/SillyTavern/blob/release/public/index.html)
-   [SillyTavern release 1.19.0 on GitHub, published 2026-09-14, latest release (checked 2026-11-16)](https://github.com/SillyTavern/SillyTavern/releases/tag/1.19.0)
-   [BroadNet Acceptable Use Policy (checked 2026-11-15)](https://www.broadnet.ai/aup)
-   [BroadNet Inference API reference (checked 2026-11-15)](https://inference.broadnet.ai/docs?utm_source=broadrouter&utm_medium=referral&utm_campaign=guides-sillytavern-custom-openai-endpoint&utm_content=sources)
