mirror of
https://github.com/ggml-org/llama.cpp.git
synced 2025-10-28 08:31:25 +00:00
chat : clarify the meaning of reasoning_format (#15408)
* chat : clarify the meaning of reasoning_format * add link to this PR
This commit is contained in:
@@ -632,7 +632,6 @@ const char * common_reasoning_format_name(common_reasoning_format format) {
|
|||||||
case COMMON_REASONING_FORMAT_AUTO: return "auto";
|
case COMMON_REASONING_FORMAT_AUTO: return "auto";
|
||||||
case COMMON_REASONING_FORMAT_DEEPSEEK: return "deepseek";
|
case COMMON_REASONING_FORMAT_DEEPSEEK: return "deepseek";
|
||||||
case COMMON_REASONING_FORMAT_DEEPSEEK_LEGACY: return "deepseek-legacy";
|
case COMMON_REASONING_FORMAT_DEEPSEEK_LEGACY: return "deepseek-legacy";
|
||||||
case COMMON_REASONING_FORMAT_GRANITE: return "granite";
|
|
||||||
default:
|
default:
|
||||||
throw std::runtime_error("Unknown reasoning format");
|
throw std::runtime_error("Unknown reasoning format");
|
||||||
}
|
}
|
||||||
|
|||||||
@@ -239,12 +239,15 @@ struct common_params_diffusion {
|
|||||||
bool add_gumbel_noise = false; // add gumbel noise to the logits if temp > 0.0
|
bool add_gumbel_noise = false; // add gumbel noise to the logits if temp > 0.0
|
||||||
};
|
};
|
||||||
|
|
||||||
|
// reasoning API response format (not to be confused as chat template's reasoning format)
|
||||||
enum common_reasoning_format {
|
enum common_reasoning_format {
|
||||||
COMMON_REASONING_FORMAT_NONE,
|
COMMON_REASONING_FORMAT_NONE,
|
||||||
COMMON_REASONING_FORMAT_AUTO,
|
COMMON_REASONING_FORMAT_AUTO, // Same as deepseek, using `message.reasoning_content`
|
||||||
COMMON_REASONING_FORMAT_DEEPSEEK_LEGACY, // Extract thinking tag contents and return as `message.reasoning_content`, or leave inline in <think> tags in stream mode
|
COMMON_REASONING_FORMAT_DEEPSEEK_LEGACY, // Extract thinking tag contents and return as `message.reasoning_content`, or leave inline in <think> tags in stream mode
|
||||||
COMMON_REASONING_FORMAT_DEEPSEEK, // Extract thinking tag contents and return as `message.reasoning_content`, including in streaming deltas.
|
COMMON_REASONING_FORMAT_DEEPSEEK, // Extract thinking tag contents and return as `message.reasoning_content`, including in streaming deltas.
|
||||||
COMMON_REASONING_FORMAT_GRANITE, // Extract thinking tag contents and return as `message.reasoning_content`, including in streaming deltas.
|
// do not extend this enum unless you absolutely have to
|
||||||
|
// in most cases, use COMMON_REASONING_FORMAT_AUTO
|
||||||
|
// see: https://github.com/ggml-org/llama.cpp/pull/15408
|
||||||
};
|
};
|
||||||
|
|
||||||
|
|
||||||
|
|||||||
@@ -1408,7 +1408,7 @@ static void test_template_output_parsers() {
|
|||||||
/* is_partial= */ false,
|
/* is_partial= */ false,
|
||||||
{
|
{
|
||||||
/* .format = */ COMMON_CHAT_FORMAT_GRANITE,
|
/* .format = */ COMMON_CHAT_FORMAT_GRANITE,
|
||||||
/* .reasoning_format = */ COMMON_REASONING_FORMAT_GRANITE,
|
/* .reasoning_format = */ COMMON_REASONING_FORMAT_DEEPSEEK,
|
||||||
}));
|
}));
|
||||||
|
|
||||||
// Test parsing tool calls
|
// Test parsing tool calls
|
||||||
|
|||||||
Reference in New Issue
Block a user