tqcq/llama.cpp
0
0
Fork 0
mirror of https://github.com/ggml-org/llama.cpp.git synced 2025-08-15 12:42:40 -04:00
Code Issues Packages Projects Releases Wiki Activity

server : allow specifying reasoning_format in HTTP request (#15238)

Browse Source
This commit is contained in:
Xuan-Son Nguyen
2025-08-11 14:48:41 +02:00
committed by GitHub
parent 27093afe78
commit 53d0a12658
7 changed files with 28 additions and 7 deletions
Download Patch File Download Diff File Expand all files Collapse all files

6
tools/server/README.md
View File

@@ -1132,6 +1132,12 @@ The `response_format` parameter supports both plain JSON output (e.g. `{"type":
`chat_template_kwargs`: Allows sending additional parameters to the json templating system. For example: `{"enable_thinking": false}`
`reasoning_format`: The reasoning format to be parsed. If set to `none`, it will output the raw generated text.
`thinking_forced_open`: Force a reasoning model to always output the reasoning. Only works on certain models.
`parse_tool_calls`: Whether to parse the generated tool call.
*Examples:*
You can use either Python `openai` library with appropriate checkpoints:

BIN
tools/server/public/index.html.gz
View File

Binary file not shown.

8
tools/server/server.cpp
View File

@@ -383,8 +383,12 @@ struct server_task {
} else {
params.oaicompat_chat_syntax.format = defaults.oaicompat_chat_syntax.format;
}
params.oaicompat_chat_syntax.reasoning_format = params_base.reasoning_format;
params.oaicompat_chat_syntax.reasoning_in_content = params.stream && (params_base.reasoning_format == COMMON_REASONING_FORMAT_DEEPSEEK_LEGACY);
common_reasoning_format reasoning_format = params_base.reasoning_format;
if (data.contains("reasoning_format")) {
reasoning_format = common_reasoning_format_from_name(data.at("reasoning_format").get<std::string>());
}
params.oaicompat_chat_syntax.reasoning_format = reasoning_format;
params.oaicompat_chat_syntax.reasoning_in_content = params.stream && (reasoning_format == COMMON_REASONING_FORMAT_DEEPSEEK_LEGACY);
params.oaicompat_chat_syntax.thinking_forced_open = json_value(data, "thinking_forced_open", false);
params.oaicompat_chat_syntax.parse_tool_calls = json_value(data, "parse_tool_calls", false);
}

1
tools/server/webui/src/utils/app.context.tsx
View File

@@ -209,6 +209,7 @@ export const AppContextProvider = ({
messages,
stream: true,
cache_prompt: true,
reasoning_format: 'none',
samplers: config.samplers,
temperature: config.temperature,
dynatemp_range: config.dynatemp_range,