proof reading

This commit is contained in:
Xuan Son Nguyen 2026-01-01 12:50:45 +01:00
parent 0f70e3e0cd
commit f9a97375cb
1 changed files with 12 additions and 12 deletions

View File

@ -1,20 +1,20 @@
# llama.cpp INI preset
# llama.cpp INI Presets
## Introduction
INI preset is a feature that was added in [PR#17859](https://github.com/ggml-org/llama.cpp/pull/17859). The goal is to allow writing reusable and sharable parameter presets in llama.cpp
The INI preset feature, introduced in [PR#17859](https://github.com/ggml-org/llama.cpp/pull/17859), allows users to create reusable and shareable parameter configurations for llama.cpp.
### Using preset on server
### Using Presets with the Server
When using multiple models on server (router mode), INI preset file can be used to configure model-specific parameters. Please refer to [server documentations](../tools/server/README.md) for more.
When running multiple models on the server (router mode), INI preset files can be used to configure model-specific parameters. Please refer to the [server documentation](../tools/server/README.md) for more details.
### Using a remote preset
### Using a Remote Preset
> [!NOTE]
>
> This feature is currently only supported via the `-hf` option
> This feature is currently only supported via the `-hf` option.
For GGUF models stored on Hugging Face, you can create a file named `preset.ini` in the root directory of the repository that contains specific configurations for the current model.
For GGUF models hosted on Hugging Face, you can include a `preset.ini` file in the root directory of the repository to define specific configurations for that model.
Example:
@ -25,16 +25,16 @@ top-k = 20
top-p = 0.95
```
For security reason, only certain options are allowed. Please refer to [preset.cpp](../common/preset.cpp) for the list of allowed options.
For security reasons, only certain options are allowed. Please refer to [preset.cpp](../common/preset.cpp) for the complete list of permitted options.
Example usage:
Provided your repo is `username/my-model-with-preset` having a `preset.ini` with the content above.
Assuming your repository `username/my-model-with-preset` contains a `preset.ini` with the configuration above:
```sh
llama-cli -hf username/my-model-with-preset
# equivalent to
# This is equivalent to:
llama-cli -hf username/my-model-with-preset \
--hf-repo-draft username/my-draft-model-GGUF \
--temp 0.5 \
@ -42,9 +42,9 @@ llama-cli -hf username/my-model-with-preset \
--top-p 0.95
```
You can also optionally override preset args by specifying them in the arguments:
You can also override preset arguments by specifying them on the command line:
```sh
# forcing temp = 0.1
# Force temp = 0.1, overriding the preset value
llama-cli -hf username/my-model-with-preset --temp 0.1
```