self-instruct

ChatGPT爆火，开启了通往AGI的关键一步，本项目旨在汇总那些ChatGPT的开源平替们，包括文本大模型、多模态大模型等，为大家提供一些便利

A collection of awesome-prompt-datasets, awesome-instruction-dataset, to train ChatLLM such as chatgpt 收录各种各样的指令数据集, 用于训练 ChatLLM 模型。

Clinically Adapted Model Enhanced from LLaMA

Unified framework for synthesizing tool-use / function-calling training data — faithful reproductions of ToolACE, ACEBench, AgentInstruct, Self-Instruct, Kimi K2 agentic & K3 KG-guided pipelines

Agentic synthetic-data generation framework inspired by Meta FAIR's Autodata / Agentic Self-Instruct.

[DEPRECATED]

Using LoRa and self-instruct to fine-tune Llama 2 7B to give you tips on Super Smash Bros. Melee

Provider-agnostic Agentic Self-Instruct SDK and CLI with OpenTelemetry trace ingestion

Budget-aware Agentic Self-Instruct pipeline for generating hard research QA data with the OpenAI API.

Fine-tuning Qwen2.5-0.5B-Instruct with QLoRA on the Self-Instruct 82K dataset. Evaluated on 252 expert-written tasks vs. OpenAI Davinci using BLEU, ROUGE-L, and BERTScore.

Vietnamese GSM8K agentic self-instruct dataset — proven to beat machine-translated data (Qwen3-4B: 81.0% vs 76.5%)