mirror of https://github.com/mudler/LocalAI.git
31 lines
1.2 KiB
Markdown
31 lines
1.2 KiB
Markdown
|
|
||
|
|
+++
|
||
|
|
disableToc = false
|
||
|
|
title = "✍️ Constrained grammars"
|
||
|
|
weight = 15
|
||
|
|
+++
|
||
|
|
|
||
|
|
The chat endpoint accepts an additional `grammar` parameter which takes a [BNF defined grammar](https://en.wikipedia.org/wiki/Backus%E2%80%93Naur_form).
|
||
|
|
|
||
|
|
This allows the LLM to constrain the output to a user-defined schema, allowing to generate `JSON`, `YAML`, and everything that can be defined with a BNF grammar.
|
||
|
|
|
||
|
|
{{% alert note %}}
|
||
|
|
This feature works only with models compatible with the [llama.cpp](https://github.com/ggerganov/llama.cpp) backend (see also [Model compatibility]({{%relref "docs/reference/compatibility-table" %}})). For details on how it works, see the upstream PRs: https://github.com/ggerganov/llama.cpp/pull/1773, https://github.com/ggerganov/llama.cpp/pull/1887
|
||
|
|
{{% /alert %}}
|
||
|
|
|
||
|
|
## Setup
|
||
|
|
|
||
|
|
Follow the setup instructions from the [LocalAI functions]({{%relref "docs/features/openai-functions" %}}) page.
|
||
|
|
|
||
|
|
## 💡 Usage example
|
||
|
|
|
||
|
|
For example, to constrain the output to either `yes`, `no`:
|
||
|
|
|
||
|
|
```bash
|
||
|
|
curl http://localhost:8080/v1/chat/completions -H "Content-Type: application/json" -d '{
|
||
|
|
"model": "gpt-4",
|
||
|
|
"messages": [{"role": "user", "content": "Do you like apples?"}],
|
||
|
|
"grammar": "root ::= (\"yes\" | \"no\")"
|
||
|
|
}'
|
||
|
|
```
|