agents

Fine-tuning strategist who owns the eval gate and method/model selection. Refuses to plan training without a baselined eval harness. Use PROACTIVELY when a user

所属插件包：llm-finetuning

Evaluation gatekeeper for fine-tuning — builds golden sets and graders, calibrates judges, baselines base models, and issues checkpoint promotion verdicts. Use

所属插件包：llm-finetuning

Fine-tuning implementation workhorse — prepares datasets, generates Unsloth-first training scripts, launches and monitors runs, and exports artifacts. Use after

所属插件包：llm-finetuning

Run the eval-gated fine-tuning lifecycle end to end — eval harness, method selection, data, environment, training, checkpoint gate, export

所属插件包：llm-finetuning

Re-gate an existing fine-tuned checkpoint against the current eval harness and export it on PROMOTE

所属插件包：llm-finetuning

Gate fine-tuned checkpoints with drift budgets, paired comparison, and forgetting checks before promotion. Use after a training run produces a checkpoint, when

所属插件包：llm-finetuning

Prepare, format, and validate datasets for supervised fine-tuning and preference training. Use when converting raw data into training format, applying chat temp

所属插件包：llm-finetuning

Build the evaluation harness that gates every fine-tuning run — golden sets, per-failure-mode graders, judge calibration, and base-model baselines. Use when sta

所属插件包：llm-finetuning

Decide whether to fine-tune at all, and route to the right method (SFT, DPO/ORPO/KTO, GRPO/RLVR, continued pretraining) and base model. Use when starting any fi

所属插件包：llm-finetuning

Train reasoning and verifiable-task behavior with GRPO and reinforcement learning from verifiable rewards (RLVR). Use when task success is algorithmically check

所属插件包：llm-finetuning

Configure LoRA and QLoRA supervised fine-tuning with current best-practice hyperparameters. Use when writing or reviewing a LoRA/QLoRA training configuration, c

所属插件包：llm-finetuning

Align a fine-tuned model with preference data using DPO, ORPO, KTO, or SimPO. Use when preference pairs or thumbs-up/down feedback exist, when choosing between

所属插件包：llm-finetuning

Export a promoted fine-tuned model in the right deployment format — merged safetensors, LoRA-only, GGUF with imatrix, or FP8. Use after a checkpoint passes prom

所属插件包：llm-finetuning

Convert evaluation traces and production logs into SFT examples and preference pairs. Use when graded traces or failure examples exist and need to become traini

所属插件包：llm-finetuning

Fine-tune vision-language models (VLMs) with supervised learning on image+text data. Use when adapting a VLM to a visual domain or task, configuring frozen-visi

所属插件包：llm-finetuning