Agent

@metaharness/weight-eft

Fine-tune cheap open-source LLMs (GLM, Qwen, DeepSeek) on your AI coding agent's successful runs with LoRA (SFT + DPO) so your model cascade escalates to expensive frontier models (GPT, Claude) less often — cutting cost-per-resolved.

Interfaces & auth

Interfaces: not recorded

Auth methods: not recorded

Verification history

Verification is automated (see methodology). Every run is stored permanently; consecutive runs with an identical result are shown as one row.