OpenHermes-2.5

← All datasets

Author: Distillio · License: unknown · Format: json

0 downloads · 0 stars · 0 mirrors · size unknown · 1 file

~1M GPT-4-generated instruction pairs deduplicated to ShareGPT/ChatML format.

Tags: sft · instruction-tuning · gpt4-distilled

Description

~1M primarily GPT-4-generated instructions, deduplicated and normalized to ShareGPT/ChatML format. No single unified license — check source datasets before commercial use.

Download

Direct download: https://huggingface.co/datasets/teknium/OpenHermes-2.5/resolve/main/openhermes2_5.json

Or via the API: curl -L -o file "https://huggingface.co/datasets/teknium/OpenHermes-2.5/resolve/main/openhermes2_5.json"

Added 2026-08-11. Platform: Distillio — the open source training data collective.