datasets
Training and evaluation data, with the modality, task and licence stated up front. Listed live from the Hugging Face Hub.
archon-sft-v1-orchestration
archon-sft-v1-orchestration
AETHER family SFT dataset — group orchestration.
Format: JSONL ChatML messages, task_type tagged, MinHash dedup applied (threshold 0.85).
Schema:
{
"messages": [{"role": "system|user|assistant", "content": "..."}],
"task_type": "function_calling|code|reasoning_cot|...",
"source_ds": "<HF dataset_id>",
"lang": "en|fr|...",
"system_source": "archon_default|overridden_from_source"
}
Generated by prepare_sft.py pipeline (2026-05-25).… See the full description on the dataset page: https://huggingface.co/datasets/jescy525/archon-sft-v1-orchestration.keural-v2-orchestration
Agent Orchestration Dataset
Status: generation complete. 50,000 / 50,000 target rows. Not yet processed — see Pipeline stage before using this for training.
File: agent_orchestration_final_50000.jsonl (597 MB, one JSON object per line, 50,000 lines).
What this is
Synthetic multi-agent conversation traces for training a model to act as an orchestrator: decompose a user task, delegate subtasks to named sub-agents, then synthesize their independent responses into one… See the full description on the dataset page: https://huggingface.co/datasets/mkd-minju/keural-v2-orchestration.agent-orchestration-patterns-v1
Agent-Orchestration-Patterns
Instruction-tuning / few-shot data for one skill: designing reliable multi-agent LLM
systems.
Each row poses a multi-agent design problem and answers with a tested pattern in a fixed
shape:
the pattern → when to use it → the anti-pattern it replaces.
These patterns were distilled from incidents and verified outcomes in a real autonomous
multi-agent run — load spikes from unscoped parallel agents, "compiled but hanging" work
passed as done, plugins… See the full description on the dataset page: https://huggingface.co/datasets/CatQualia/agent-orchestration-patterns-v1.