**Five common instruction-dataset formats** show up in fine-tuning workflows, and each has a recognisable on-disk shape.
Alpaca uses flat records. Each row is one example with three fields: instruction (the task), input (optional context, often empty), and output (the target response). It is the simplest format and a common starting point for early SFT.
ShareGPT stores multi-turn conversations as a list per row. Each row has a conversations field that contains a list of turns, each turn tagged with from (the speaker, usually human or gpt) and value (the text). The format originated from exported ChatGPT-style transcripts.
OpenAssistant or OASST stores conversations as a tree spread across many rows. Each row is a single message with role, parent_id, message_id, and rank fields. You reconstruct a full thread by walking parent links. The tree shape supports multiple alternative replies branching from any message.
FLAN uses templated task pairs. Each row has input_text and target_text fields, with the task instruction baked into the input by a prompt template. It comes from the FLAN collection of academic NLP datasets reformatted for instruction tuning.
OpenAI Chat-JSONL is the format the hosted fine-tuning API consumes. Each line is a JSON object with a messages field that contains a list of turns tagged with role (system, user, or assistant) and content.
**Five common instruction-dataset shapes**, each with a recognisable on-disk signature.
Alpaca: flat records, one example per row. Each row has three fields, instruction (the task), input (optional context, often empty), and output (the target response). Originated with the Stanford Alpaca release in 2023 and remains a common starter format for simple SFT.
ShareGPT: multi-turn conversation per row. Each row carries a conversations field holding a list of turns, each turn tagged with from for the speaker (usually human or gpt) and value for the text. The format came from exported ChatGPT-style transcripts and is still common in community datasets.
OpenAssistant or OASST: tree across many rows. Each row is one message with role, parent_id, message_id, and rank fields. Threads are reconstructed by walking parent links. The tree shape is essential because OASST captures multiple alternative replies branching from any node, which is needed for preference data.
FLAN: templated task pairs. Each row has input_text and target_text (or inputs and targets) fields. The task instruction is baked into input_text by a prompt template drawn from the FLAN collection of NLP datasets. It is broad but flat, no multi-turn context.
OpenAI Chat-JSONL: the canonical hosted-FT format. One JSON object per line with a messages array of turns, each turn carrying role (system, user, assistant, or tool) and content. The format is strict and what the OpenAI fine-tuning API consumes directly.
Quick reference
TL;DR
Alpaca is flat instruction, input, output, ShareGPT is a conversation list with from tags, OASST is a parent_id message tree, FLAN is input_text-target_text pairs, OpenAI Chat-JSONL is the role-tagged messages array.
Imagine five different ways to record a tutoring conversation. Alpaca writes one neat card with the question, optional context, and the answer. ShareGPT writes a small script of who said what in order. OASST is more like a family tree of replies where any line can branch into several follow-ups. FLAN fills out a worksheet with a prompt and the expected response. OpenAI Chat-JSONL is a tidy diary where every entry tags the speaker as system, user, or assistant. Each format suits a different purpose, and recognising the shape on disk tells you what the dataset was built for.
Everything you need to truly understand this topic: intuition, mechanics, step by step explanation, code, formulas, and worked example. Click to expand.
Everything you need to truly understand this topic: intuition, mechanics, step by step explanation, code, formulas, and worked example.
Pretraining a base model rather than instruction-tuning These formats encode instructions and conversations, pretraining corpora are raw text or document streams, the structured shape adds no value and just complicates the loader.
Preference optimisation with DPO or similar where chosen and rejected pairs are needed These formats encode single responses or unbranched conversations, DPO needs explicit pair structure with chosen and rejected fields, the standard SFT shapes do not capture that.
Tool-calling data with structured function calls and tool responses Most of these formats predate tool calling, only the OpenAI Chat-JSONL format (with tool_calls and tool roles) and ShareGPT extensions support the structure cleanly.
Quick scan · 2–4 min
Everything important, quickly.
4 min: topology sort first (three flat, one list, one tree) + field-name vocabulary per format + how to disambiguate ShareGPT from OpenAI Chat-JSONL + why OASST is a tree + format adapters in modern stacks.
Real-world examples
Real products, models, and research that use this idea.
1Hugging Face's tulu-v3 mix in 2026 converts source datasets including Alpaca, ShareGPT, and OASST descendants into a unified ShareGPT-shape JSONL for community fine-tuning.
2OpenAI's hosted fine-tuning for gpt-5.5 accepts only the Chat-JSONL format, so most other formats must be transformed before upload.
3Axolotl and LLaMA-Factory both ship dataset-format adapters for Alpaca, ShareGPT, OASST, FLAN, and Chat-JSONL, automatically applying the right chat template per format.
4The OpenAssistant project's released data uses the tree format on Hugging Face Hub, and most downstream uses flatten preferred-rank paths into ShareGPT-shape conversations for training.
Interviewer follow-ups
What an interviewer would ask next. Try answering before peeking at the approach.
QWhy does OASST use a tree structure instead of just storing flat conversations?
Sign in or go Pro to see the approach for this follow-up.
QWhat information is lost when converting an OASST tree to a ShareGPT-shape conversation list?
Sign in or go Pro to see the approach for this follow-up.
QWhy do OpenAI Chat-JSONL roles include system and tool when ShareGPT mostly does not?
Sign in or go Pro to see the approach for this follow-up.
Don't say this
Red flags and common mistakes that signal junior thinking. Click to expand.
Most common mistake
Confusing ShareGPT and OpenAI Chat-JSONL because both encode conversations. The giveaway is the field names, ShareGPT uses from and value with informal speaker tags, OpenAI uses role and content with strict role names.