🇰🇿 Kazakh API Discovery and Tool Retrieval Dataset
Dataset Summary
Kazakh API Discovery and Tool Retrieval Dataset is a Kazakh-language dataset designed for training and evaluating Large Language Models (LLMs) in agentic AI workflows that require API discovery, tool documentation retrieval, function calling, and multi-step tool execution.
The dataset focuses on scenarios where the assistant must first inspect or retrieve API documentation before calling the appropriate tool. This makes it useful for training agents that do not simply call tools directly, but first verify the required parameters, understand the tool schema, and then execute the task with correctly structured arguments.
📊 Dataset Statistics
General Metrics
| Metric | Count |
|---|---|
| Total Samples | 2,552 |
| Total Words (approx.) | 1,159,748 |
| Avg. Words per Sample | 454 |
Word Count Distribution Per Field
The following table details the distribution of word counts across different fields in the dataset.
| Field | Mean | Median | Min | Max | Total Words |
|---|---|---|---|---|---|
| answers | 58.1 | 55.0 | 18 | 173 | 148,178 |
| category | 1.0 | 1.0 | 1 | 1 | 2,552 |
| difficulty | 1.0 | 1.0 | 1 | 1 | 2,552 |
| id | 1.0 | 1.0 | 1 | 1 | 2,552 |
| query | 15.6 | 15.0 | 4 | 51 | 39,739 |
| tools | 128.0 | 128.0 | 72 | 241 | 326,635 |
| turns | 249.8 | 244.0 | 122 | 574 | 637,540 |
Dataset Structure
Each dataset instance represents a complete API-discovery and tool-execution workflow. A sample usually includes a Kazakh user request, several available tools, expected documentation-retrieval calls, executable tool calls, simulated tool responses, and a final answer.
Data Fields
id: A unique identifier for the sample.query: The original user request in Kazakh. The query may combine several tasks, such as making a reservation, searching recent news, and retrieving general web information.category: The task category. For example,04_api_discovery_retrievalindicates that the sample focuses on retrieving API documentation before using tools.tools: A list of available tools. Each tool contains:name: the tool name;description: the tool’s purpose;parameters: the declared argument schema;mock_response: the expected response format, when available.
difficulty: The difficulty level of the sample.answers: The expected sequence of tool calls. This field may include both documentation retrieval calls, such asdocs.retrieve, and execution calls, such asrestaurant.reserve,news.search, orweb.search.turns: The complete interaction trajectory, including:- user request;
- assistant planning;
- API documentation retrieval;
- tool calls;
- mock tool outputs;
- final synthesized answer.
Data Instance
Below is one representative example from the dataset. It demonstrates API documentation retrieval before executing multiple tools. :contentReference[oaicite:0]{index=0}
{
"id": "04_api_discovery_0226_2010",
"query": "Ертең 19:30-ға 4 адамға \"R-102\" мейрамханасына бронь жасап, соңғы жаңалықтардан \"Астана туризм\" туралы мақалалар тап та, вебтен сол тақырып бойынша қысқаша ақпарат ізде.",
"category": "04_api_discovery_retrieval",
"tools": [
{
"name": "restaurant.reserve",
"description": "Make restaurant reservation",
"parameters": {
"restaurantId": {
"type": "string",
"description": "Restaurant ID",
"required": true
},
"date": {
"type": "string",
"description": "Reservation date YYYY-MM-DD",
"required": true
},
"time": {
"type": "string",
"description": "Time HH:MM",
"required": true
},
"guests": {
"type": "int",
"description": "Number of guests",
"required": true
}
},
"mock_response": {
"reservation_id": "",
"status": "",
"restaurant": "",
"time": ""
}
},
{
"name": "news.search",
"description": "Search recent news articles",
"parameters": {
"query": {
"type": "string",
"description": "Search topic",
"required": true
},
"language": {
"type": "string",
"description": "Language code",
"required": false
},
"pageToken": {
"type": "string",
"description": "Pagination token",
"required": false
}
},
"mock_response": {
"articles": [
{
"source": "",
"title": "",
"date": "",
"url": ""
}
]
}
},
{
"name": "web.search",
"description": "Search the web for information",
"parameters": {
"query": {
"type": "string",
"description": "Search query",
"required": true
},
"limit": {
"type": "int",
"description": "Number of results",
"required": false
}
},
"mock_response": {
"results": [
{
"title": "",
"snippet": "",
"url": ""
}
]
}
},
{
"name": "docs.retrieve",
"description": "Get API documentation for a service",
"parameters": {
"service": {
"type": "string",
"description": "Service name",
"required": true
},
"function": {
"type": "string",
"description": "Function name",
"required": true
}
},
"mock_response": {
"parameters": {},
"required": []
}
}
],
"difficulty": "easy",
"answers": [
{
"name": "docs.retrieve",
"arguments": "{\"service\": \"restaurant\", \"function\": \"reserve\"}"
},
{
"name": "restaurant.reserve",
"arguments": "{\"restaurantId\": \"R-102\", \"date\": \"2026-03-05\", \"time\": \"19:30\", \"guests\": 4}"
},
{
"name": "docs.retrieve",
"arguments": "{\"service\": \"news\", \"function\": \"search\"}"
},
{
"name": "news.search",
"arguments": "{\"query\": \"Астана туризм\", \"language\": \"kk\", \"pageToken\": \"page12345nextToken\"}"
},
{
"name": "docs.retrieve",
"arguments": "{\"service\": \"web\", \"function\": \"search\"}"
},
{
"name": "web.search",
"arguments": "{\"query\": \"Астана туризм 2026 негізгі көрсеткіштер\", \"limit\": 3}"
}
],
"turns": [
{
"role": "user",
"content": "Ертең 19:30-ға 4 адамға \"R-102\" мейрамханасына бронь жасап, соңғы жаңалықтардан \"Астана туризм\" туралы мақалалар тап та, вебтен сол тақырып бойынша қысқаша ақпарат ізде."
},
{
"role": "assistant",
"content": "Алдымен бронь жасаймын, кейін жаңалықтар мен вебтен сұралған тақырып бойынша нәтижелерді жинақтаймын.",
"meta": {
"plan": "restaurant.reserve + news.search + web.search analysis api with correct params, then fetching",
"tags": [],
"difficulty": "hard"
}
},
{
"role": "assistant",
"content": "",
"tool_call": {
"name": "docs.retrieve",
"arguments": "{\"service\": \"restaurant\", \"function\": \"reserve\"}"
}
},
{
"role": "tool",
"content": "{\"parameters\": {\"restaurantId\": \"string\", \"date\": \"string\", \"time\": \"string\", \"guests\": \"int\"}, \"required\": [\"restaurantId\", \"date\", \"time\", \"guests\"]}"
},
{
"role": "assistant",
"content": "",
"tool_call": {
"name": "restaurant.reserve",
"arguments": "{\"restaurantId\": \"R-102\", \"date\": \"2026-03-05\", \"time\": \"19:30\", \"guests\": 4}"
}
},
{
"role": "tool",
"content": "{\"reservation_id\": \"RES-8F3K21\", \"status\": \"confirmed\", \"restaurant\": \"R-102\", \"time\": \"19:30\"}"
},
{
"role": "assistant",
"content": "",
"tool_call": {
"name": "docs.retrieve",
"arguments": "{\"service\": \"news\", \"function\": \"search\"}"
}
},
{
"role": "tool",
"content": "{\"parameters\": {\"query\": \"string\", \"language\": \"string\", \"pageToken\": \"string\"}, \"required\": [\"query\"]}"
},
{
"role": "assistant",
"content": "",
"tool_call": {
"name": "news.search",
"arguments": "{\"query\": \"Астана туризм\", \"language\": \"kk\", \"pageToken\": \"page12345nextToken\"}"
}
},
{
"role": "tool",
"content": "{\"articles\": [{\"source\": \"Tengrinews\", \"title\": \"Астанада туризм көрсеткіші өсті: жаңа бағыттар мен іс-шаралар\", \"date\": \"2026-03-03\", \"url\": \"https://example.com/news/astana-tourism-growth\"}]}"
},
{
"role": "assistant",
"content": "",
"tool_call": {
"name": "docs.retrieve",
"arguments": "{\"service\": \"web\", \"function\": \"search\"}"
}
},
{
"role": "tool",
"content": "{\"parameters\": {\"query\": \"string\", \"limit\": \"int\"}, \"required\": [\"query\"]}"
},
{
"role": "assistant",
"content": "",
"tool_call": {
"name": "web.search",
"arguments": "{\"query\": \"Астана туризм 2026 негізгі көрсеткіштер\", \"limit\": 3}"
}
},
{
"role": "tool",
"content": "{\"results\": [{\"title\": \"Astana Tourism: Visitor stats and key attractions (2026 overview)\", \"snippet\": \"Қалаға келушілер саны, негізгі көрікті орындар және маусымдық іс-шаралар туралы қысқаша шолу.\", \"url\": \"https://example.com/web/astana-tourism-2026\"}]}"
},
{
"role": "assistant",
"content": "Бронь дайын: R-102, 2026-03-05 күні 19:30, 4 адам, мәртебе confirmed (ID: RES-8F3K21). Жаңалық нәтижесі: Tengrinews — «Астанада туризм көрсеткіші өсті: жаңа бағыттар мен іс-шаралар» (2026-03-03): https://example.com/news/astana-tourism-growth. Веб іздеу нәтижесі: «Astana Tourism: Visitor stats and key attractions (2026 overview)» — қысқаша шолу: https://example.com/web/astana-tourism-2026. (Дерек көзі: restaurant.reserve + news.search + web.search)"
}
]
}
Funding
This dataset was developed as part of the project funded by the Ministry of Science and Higher Education of the Republic of Kazakhstan under Grant No. BR24993001, “Creation of a Large Language Model (LLM) to Support the Kazakh Language and Advance Technological Development.”
Citation
If you use this dataset in your research, please cite the following article:
APA
Kadyrbek, N., Tuimebayev, Z., Mansurova, M., & Viegas, V. (2025). The development of small-scale language models for low-resource languages, with a focus on Kazakh and direct preference optimization. Big Data and Cognitive Computing, 9(5), 137. https://doi.org/10.3390/bdcc9050137
BibTeX
@article{kadyrbek2025development,
title = {The Development of Small-Scale Language Models for Low-Resource Languages, with a Focus on Kazakh and Direct Preference Optimization},
author = {Kadyrbek, Nurgali and Tuimebayev, Zhanseit and Mansurova, Madina and Viegas, Vitor},
journal = {Big Data and Cognitive Computing},
volume = {9},
number = {5},
pages = {137},
year = {2025},
publisher = {MDPI},
doi = {10.3390/bdcc9050137},
url = {https://www.mdpi.com/2504-2289/9/5/137}
}
- Downloads last month
- 13
