Datasets:

Offlin33er
/

glm-5.3-flash-function-calling

tools listlengths 1 4	messages listlengths 2 7	category stringclasses 5 values
[ { "type": "function", "function": { "name": "create_calendar_event", "description": "Create a new event on the user's calendar with the specified title, date, time, duration, and optional location.", "parameters": { "type": "object", "properties": { "title": { ...	[ { "role": "system", "content": "You are a scheduling assistant with access to the user's calendar. Use the available tools to help the user manage their schedule.", "tool_calls": null, "name": null }, { "role": "user", "content": "Can you put a dentist appointment on my calendar for Marc...	single_call
[ { "type": "function", "function": { "name": "create_reminder", "description": "Create a new reminder with a title and a due date/time. Optionally include notes and a priority level.", "parameters": { "type": "object", "properties": { "title": { "type":...	[ { "role": "system", "content": "You are a personal assistant that helps manage to-dos and reminders. Use the available tools when needed to complete the user's request.", "tool_calls": null, "name": null }, { "role": "user", "content": "Can you remind me to call the dentist tomorrow at 9...	single_call
[ { "type": "function", "function": { "name": "get_file_metadata", "description": "Retrieve metadata for a single file, including size, creation date, permissions, and checksum.", "parameters": { "type": "object", "properties": { "title": null, "date": nul...	[ { "role": "system", "content": "You are a file management assistant with access to tools for searching and inspecting files. Use the tools whenever the user requests file information.", "tool_calls": null, "name": null }, { "role": "user", "content": "Can you check the metadata for /home...	parallel_calls
[ { "type": "function", "function": { "name": "search_tracks", "description": "Search the music library for tracks matching a title, artist, or album, with optional filters for genre and release year.", "parameters": { "type": "object", "properties": { "title": null...	[ { "role": "system", "content": "You are TuneAssist, a helpful assistant for managing the user's personal music library. You can search for tracks and create playlists. Always confirm details before creating playlists.", "tool_calls": null, "name": null }, { "role": "user", "content": "Ca...	multi_turn
[ { "type": "function", "function": { "name": "search_files", "description": "Search for files and folders matching a pattern within a specified directory, optionally filtering by file type.", "parameters": { "type": "object", "properties": { "title": null, ...	[ { "role": "system", "content": "You are a file management assistant. Use the available tools to help the user locate, organize, and manage files on their system.", "tool_calls": null, "name": null }, { "role": "user", "content": "Can you find all the PDF invoices in my Downloads folder t...	single_call
[ { "type": "function", "function": { "name": "create_calendar_event", "description": "Creates a new event on the user's calendar with a title, start time, end time, and optional location.", "parameters": { "type": "object", "properties": { "title": { "t...	[ { "role": "system", "content": "You are a scheduling assistant with access to the user's calendar tools. Answer questions helpfully and only use tools when needed to manage events or check availability.", "tool_calls": null, "name": null }, { "role": "user", "content": "How many days are...	no_tool
[ { "type": "function", "function": { "name": "search_products", "description": "Search the product catalog by keyword with optional filters for category, price range, and sorting.", "parameters": { "type": "object", "properties": { "title": null, "date": ...	[ { "role": "system", "content": "You are ShopAssist, an e-commerce shopping assistant. Help users find products, compare options, and manage their orders using the available tools.", "tool_calls": null, "name": null }, { "role": "user", "content": "Can you find me a coffee maker under $80...	single_call
[ { "type": "function", "function": { "name": "get_monthly_spending_breakdown", "description": "Retrieves a categorized breakdown of the user's spending for a given month and year, based on linked bank and credit card accounts.", "parameters": { "type": "object", "properties"...	[ { "role": "system", "content": "You are Penny, a personal finance assistant with access to the user's linked accounts. Help the user understand and manage their money. When the user asks about their finances, use the available tools to fetch accurate data before answering.", "tool_calls": null, "nam...	single_call
[{"type":"function","function":{"name":"set_thermostat_mode","description":"Set the thermostat to a (...TRUNCATED)	[{"role":"system","content":"You are HomeSense, a smart home assistant. You help users control therm(...TRUNCATED)	multi_turn
[{"type":"function","function":{"name":"search_movies_by_title","description":"Search for movies by (...TRUNCATED)	[{"role":"system","content":"You are a helpful entertainment assistant. Use the available tools to l(...TRUNCATED)	parallel_calls

End of preview. Expand in Data Studio

GLM-5.3-Flash function-calling (synthetic)

500 synthetic function-calling training samples generated with zai-org/GLM-5.3-Flash via HF Inference Providers (auto-routing with novita / together / fireworks fallback), 2026-09-21.

Schema

Each row:

messages — chat in OpenAI tool-use format (system / user / assistant); the final assistant message either carries tool_calls (with JSON-string arguments) or is a plain-text answer
tools — 1–4 tool schemas in OpenAI function format (type, function{name, description, parameters})
category — one of:
- single_call — exactly one tool call
- parallel_calls — 2–4 independent calls in one response
- multi_turn — multi-turn with tool-result turns
- error_handling — call with plausibly failing arguments
- no_tool — tempting but irrelevant tools; assistant answers directly

Domains span weather, travel, finance, e-commerce, smart home, file management, and ~14 more. Every sample passed a validator: tool-call names resolve to declared tools, arguments parse as JSON objects, last message is assistant-role, message roles valid.

Generation

Single-shot JSON generation per sample (temperature=0.9, seeded), 10-way concurrency, per-sample validation with provider fallback. Generation script pattern available in the conversation that produced this dataset. Costs: ~7 min on cpu-basic (Jobs) + provider inference.

Downloads last month: 17