Superminaren/llama.cpp
1
0
Fork 0
mirror of https://github.com/ggml-org/llama.cpp.git synced 2026-09-27 21:46:57 +02:00
Code Issues Packages Projects Releases Wiki Activity

chat : add dedicated Ling 3.0 (Bailing V3) parser (#28682)

Browse Source 
* chat: add dedicated Ling 3.0 (Bailing V3) parser

Ling 3.0 Flash templates pre-open the think block in the generation
prompt, so the model never emits an opening <think>, and a tool call can
arrive before any </think>. The generated autoparser terminated reasoning
only at the close tag, which classified such tool calls entirely as
reasoning_content: clients received content="" with no tool_calls and
agent loops died as reasoning-only turns.

Adds a specialized parser that terminates reasoning at the think close
tag or at a <tool_call> start, mirroring the hand-written Qwen3-Coder and
Kimi K3 parsers and the reference vLLM/SGLang Ling3 parser (which treats
<tool_call> as an implicit reasoning terminator). Detection is gated on
the <role>...</role> section markers, unique to this family among the
tagged-argument templates.

Adds the Ling 3.0 Flash chat template and tests covering the
unclosed-think tool call (full parse and streaming), healthy closed-think
paths, trailing prose, parallel calls, marker-like strings in argument
values, string-union and non-string argument types, and
reasoning_format=none.

Assisted-by: Kimi Code

* tests : move Ling 3.0 test

---------

Co-authored-by: aetherbird <[email protected]>
Co-authored-by: Alde Rojas <[email protected]>
This commit is contained in:
Toby
2026-09-19 18:35:44 -05:00
committed by GitHub
co-authored by aetherbird Alde Rojas
parent e613ef2c81
commit 59657a613a
6 changed files with 543 additions and 0 deletions
Download Patch File Download Diff File Expand all files Collapse all files
common/chat.cpp
+8
View File
@@ -1133,6 +1133,14 @@ std::optional<common_chat_params> common_chat_try_specialized_template(
return common_chat_params_init_kimi_k3(tmpl, params);
}
// Ling 3.0 / Bailing V3 - <role>X</role> sections with <arg_key>/<arg_value> tagged
// tool calls. <role> sections are unique to this family among the tagged-arg templates.
if (src.find("<role>ASSISTANT</role>") != std::string::npos &&
src.find("<arg_key>") != std::string::npos) {
LOG_DBG("Using specialized template: Ling 3.0 (Bailing V3)\n");
return common_chat_params_init_ling3(tmpl, params);
}
// Cohere2 MoE / North Code - marker-wrapped format with <|START_TEXT|> content and
// <|START_ACTION|> JSON tool calls. <|START_TEXT|> is unique to this template (the older
// Command-R templates use <|START_RESPONSE|>).