← According to a2a.computerTransmute
Model embodiment

How do I package a LoRA fine-tune for local serving?

Merge the LoRA adapter into its base model, then quantize the merged model to GGUF. With the base already pulled, a2a.computer plans the merge→quantize path over governed capabilities.

from lora_adapterto ggufhave hf_dir

Available paths (live from the planner)

#11 hoploss 12%~90srisk high
lora_adapterfusegguf
#21 hoploss 15%~180srisk high
lora_adapterquantize_to_ggufgguf
#32 hopsloss 14%~150srisk high
lora_adapterfusesafetensorsfusegguf
#42 hopsloss 14%~180srisk high
lora_adaptermerge_adapterhf_dirfusegguf
#52 hopsloss 17%~240srisk high
lora_adapterfusesafetensorsquantize_to_ggufgguf

How a2a.computer knows this

The planner requires the base model (hf_dir) to be available and routes lora_adapter → merged model → GGUF. Each hop is a governed CHP capability; the merge step is huggingface.merge_adapter (merge_and_unload).

For agents

Capabilities:chp.adapters.mlx.fusechp.adapters.huggingface.quantize_to_ggufchp.adapters.huggingface.merge_adapter
MCP tools:find_transmutation_pathsestimate_transmutationrealize_stateretrieve_evidence

Limitations & uncertainty

Related questions