A tool-calling fine-tune of LiquidAI/LFM3-1.7B on the claudecode split of saidutta69/fable-5.1-premium-v2`. It teaches the 1.7B LFM3 hybrid to emit Claude Code tool calls — Bash, Read, Write, Edit, Glob, Grep, WebFetch, WebSearch, TodoWrite, TaskCreate, TaskUpdate, TaskOutput, ToolSearch — with valid tool names and typed arguments. The base model is already agentic; this checkpoint is tuned for schema fidelity inside a Claude-Code-style harness. — Running a Claude-Code-style tool protocol on a local 1.7B model — Q4KM is 1.10 GB. — Harnesses where an invalid tool name breaks the loop. The base model invents a tool on ~23% of parsed calls; this checkpoint does not. — Not a general-purpose LFM3 upgrade. Free-form text modelling shifted; use the base for general writing. The dataset’s fablecode split is unusable for argument fidelity: 203,750 of 203,794 tool calls carry whole-value placeholders such as and (44 clean calls total). The claudecode split carries real values (tasks/SEEDSPEC.md`, real shell commands), so training used that split. The schemas in this dataset are reconstructed, not original Claude Code schemas. They are close enough for the protocol, but do not assume…
Модальности:
Генерация текста
Области применения:
Генерация кода Следование инструкциям Диалог / чат Вызов функций (Tool use) Агенты
Языки программирования:
Bash/Shell
Задача: Генерация текста
Автор: saidutta69
Теги: gguf, lfm3, lfm, liquid-ai, liquid, tool-calling, function-calling, tool-use
Лайков: 3 | Загрузок: 1,505
Описание основано на материалах HuggingFace. Перевод выполнен автоматически.
