Tuning Fork
Copyright 2026 Heyuan Huang

Licensed under the Apache License, Version 2.0 (see LICENSE).

Tuning Fork follows, or includes, work from these projects, each under the MIT License:

llama.cpp — https://github.com/ggml-org/llama.cpp
Copyright (c) 2023-2026 The ggml authors
  src/tuningfork/backends/llamacpp/anthropic.py follows, rule for rule, how llama.cpp's server
  converts an Anthropic Messages request into a chat-completions one
  (tools/server/server-chat.cpp, build b11108). Three modules in src/tuningfork/backends/formats/
  follow how llama.cpp reads a model's output, at the same build: gemma4.py its parser of
  Gemma 4's (common/parsers/gemma4.cpp); qwen3_coder.py its parser of Qwen3.5's and
  Qwen3-Coder's (common/parsers/qwen3-coder.cpp), with the types it reads a call's values as
  (common/json-schema.cpp); and qwen_json.py the parser its autoparser builds for Qwen3's and
  Qwen2.5's (common/chat-diff-analyzer.cpp, common/chat-auto-parser-generator.cpp), with its
  grammar's JSON rules (common/peg-parser.cpp). None copies its code.

Hermes Agent — https://hermes-agent.nousresearch.com
Copyright (c) 2025 Nous Research
  examples/hermes-design-md/design-md/ is Hermes Agent's own design-md skill, as Hermes 0.18.2
  ships it. Its licence is in examples/hermes-design-md/LICENSE-hermes.
