feat: render GGUF Jinja template via minijinja crate

- Replaced manual prompt building with GGUF's chat template rendered through minijinja (Rust Jinja2 engine)
- Simplified template: removed multi-step tool detection, reasoning extraction (not needed at template level)
- Added reasoning_content separation: clean_text() returns (reasoning, answer) tuple
- Added reasoning_content field to ResponseMessage and SseDelta for OpenAI-compatible output
- Embedded template at build time via include_str! from templates/chat_template.jinja
- Built-in support for enable_thinking, tool_definitions, tool_calls, tool_response

This commit is contained in:

Asep Haryana

2026-07-26 18:21:35 +07:00

parent 9a63ff1601

commit 6ff31b5f62

6 changed files with 183 additions and 55 deletions

									
										Cargo.toml
									
		+1
		
												View File
												
				@@ -6,6 +6,7 @@ edition = "2021"

				[dependencies]

				# LLM inference

				llama-cpp-2 = "0.1"

				minijinja = "2"

				# HTTP server

				axum = { version = "0.8", features = ["json"] }