MillBill: A Machining Agent with Geometric Feedback
Abstract
Turning a CAD model into a machining program requires both process knowledge and geometric precision. Language models can generate such workflows by writing geometry and toolpath code for each part, but this makes numerical execution part of the model's task. We introduce MillBill, an agent that separates machining decisions from their geometric execution and verification. Given a boundary-representation part, stock, and material, a locally hosted, mid-sized open model proposes operations and cutting parameters for indexed three-axis machining. A geometric harness generates toolpaths, simulates material removal, and returns feedback for plan refinement. Setup planning accounts for what the available cutters can physically reach, rather than visibility alone. We evaluate the resulting programs by the fraction of the target surface area within the specified tolerance after simulated machining. With Qwen3.8-27B, MillBill outperforms frontier-model approaches that construct their own geometry and toolpath workflows on the evaluated machining tasks. Code: https://github.com/kulibinai/millbill
Community
Can a language model program a CNC machine without computing a single coordinate? ๐ ๏ธ
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.
๐ 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%.
๐ฆ Open release: code, plus 4,294 GrabCAD parts paired with milling stocks and a simulator that scores any planner's G-code.
๐ป https://github.com/kulibinai/millbill
๐ค https://huggingface.co/datasets/kulibinai/millbill
This is an automated message from the Librarian Bot. I found the following papers similar to this paper.
The following papers were recommended by the Semantic Scholar API
- Geo2Gcode: From 3D Geometry to CNC Programs through Structured Machining Primitives (2026)
- CNCGEN: A Dataset and Framework for Machining Process Planning and Toolpath Generation from B-rep Models (2026)
- CADFather: Autonomous CAD Reconstruction through Coordinated Tool Use (2026)
- Parametric CAD Bench v3: Evaluating Agentic Parametric CAD Authoring from Text and Engineering Drawings (2026)
- StepCAD: Mesh-to-CAD Code Generation via LLM Policy and Geometry-Guided Search (2026)
- SubDGuide: A Modeler-Inspired Agentic Workflow for Mesh-to-SubD Reconstruction (2026)
- USDCraft: Geometrically Grounded Programmatic Modeling of Articulated 3D Assets for Simulation (2026)
Please give a thumbs up to this comment if you found it helpful!
If you want recommendations for any Paper on Hugging Face checkout this Space
You can directly ask Librarian Bot for paper recommendations by tagging it in a comment: @librarian-bot recommend
Separating the machining decisions from the geometry that carries them out is the right split. In a shop, the machinist decides the order of cuts and the jig holds the line. Planning around what the cutter can physically reach, and not just what's visible, is the detail a real setup turns on. I'd like to see the same approach tried on timber, where the stock isn't uniform. That's the material we care about at ML Systems.
Models citing this paper 0
No model linking this paper
Datasets citing this paper 1
Spaces citing this paper 0
No Space linking this paper
Collections including this paper 0
No Collection including this paper