MillBill:具幾何回饋的加工智能體
MillBill: A Machining Agent with Geometric Feedback
MillBill 將 B-rep 零件、毛坯及材料轉換為索引式三軸銑削的 G-code,並以幾何回饋協助模型修訂加工計劃。在100個 GrabCAD 零件的模擬評分中,本地運行的開源 Qwen3.8-27B 令97.4%的目標表面偏差維持在0.1 mm內;GPT-6 Astra、Claude Opus 5.5及 CNC-Net 分別達到94.3%、83.0%及42.3%。
Can a language model program a CNC machine without computing a single coordinate? 🛠️
MillBill turns a B-rep part, its stock and its material into G-code for indexed 3-axis milling. The LLM only makes the machining decisions: operations, tools, cutting parameters. A geometric harness does the geometry: it recognises features, chooses setups by what the cutters can actually reach (not just by visibility), generates every toolpath, simulates material removal and returns per-feature deviations so the model can refine the plan.
📊 On 100 GrabCAD parts, every program is scored by simulation. With the open Qwen3.8-27B running locally, MillBill leaves 97.4% of the target surface within 0.1 mm. The best frontier code-writing agent (GPT-6 Astra) reaches 94.3%, Claude Opus 5.5 83.0% and CNC-Net 42.3%.
📦 Open release: code, plus 4,294 GrabCAD parts paired with milling stocks and a simulator that scores any planner's G-code.
💻 https://github.com/kulibinai/millbill
🤗 https://huggingface.co/datasets/kulibinai/millbill
來源:HuggingFace Daily Papers(社區熱門論文) · huggingface.co