llama.cpp
`server`: fix tool-call of DeepSeek R1 Qwen, return reasoning_content (Command 7RB & DeepSeek R1) unless `--reasoning-format none`
#11607
Merged
Overview
Commits
94
Changes
View On GitHub
  • common
    • File
      arg.cpp
    • File
      chat.cpp
    • File
      chat.hpp
    • File
      common.h
    • File
      sampling.cpp
  • examples/server
    • File
      README.md
    • File
      server.cpp
    • tests
      • unit
        • File
          test_tool_call.py
      • File
        utils.py
    • File
      utils.hpp
  • models/templates
    • File
      README.md
    • File
      deepseek-ai-DeepSeek-R1-Distill-Llama-8B.jinja
    • File
      deepseek-ai-DeepSeek-R1-Distill-Qwen-32B.jinja
    • File
      llama-cpp-deepseek-r1.jinja
  • scripts
    • File
      get_chat_template.py
  • src
    • File
      llama-grammar.cpp
  • tests
    • File
      test-chat.cpp

Loading comments...

FAQ Terms Privacy Refunds Impressum