Hugging Face ships a Pi-agent LoRA for Qwen3.8-27B, cutting tool-call flailing
Disastrous-Work-1632 · reddit · 2026-11-17
Aritra from Hugging Face announced the latest harness adapter: a LoRA trained on Qwen3.8-27B for the Pi agent. The idea is fine-tuning adapters so models behave more efficiently inside a specific harness — the demo image shows that on the same SWE-bench Pro task, the base model keeps searching and reading while the pi adapter locates the function, edits it and runs tests with a single tool call.
Full reproducible setup: hf download the Qwen3.8-27B Q80 GGUF and the pi-lora-F16 adapter, serve via llama serve with the LoRA attached (thinking disabled, 64K context), then pi install the pi-llama plugin. The team shares ongoing experiments at hf.co/harness-hub.
More from coding & agent
- Stitch embeds AI agents in banking: the safety story is harnesses, not hack-proofing — alex_verem · 2026-11-17
- Agent Identity 101: one sequence diagram explains delegated, agent, and MCP auth patterns — adnan_hashmi · 2026-11-17
- How Comet improved a financial incident remediation agent's score from 0.34 to 0.57 with Opik — dl_weekly · 2026-11-17
- Jason Weston: Harness Optimization Hits 56.0, Distillation Adds 5.5 More — jaseweston · 2026-11-17
- Meta's Self-Improving Memory: agents discover their own memory skills and distill them into weights — jaseweston · 2026-11-17
- Every shows Claude Fable 5.2 building interactive games — slow but token-hungry — every · 2026-11-17