FrontierCAD Assemblies
Agents place supplied CAD parts into 16 assemblies. The score is the published mean assembly geometry score.
One track of FrontierCAD. The FrontierCAD record keeps lab scores that combine tracks.
- Made by
- Normal Research
- Released
- October 2026
- Kind
- CAD models
- Level
- Expert
- Scored
- Checked automatically
- Answers
- Agent task
- Tasks
- Kept private
- Strongest model tested
- GPT-6 Astra #3 on the capability index
- Last model added
- September 2026
- Scores refreshed
- November 12th, 2026
GPT-6 Astra leads with 0.71 among models released by September 2026
Best score per entry from the benchmark's own leaderboard, read on November 12th, 2026. Where a model ran with several settings or harnesses, we keep its best. Names in gray aren't in our model list: fine-tunes, agent systems, products and older models. They don't count toward leaders.
The newest model with a score here came out in September 2026, and nobody has tested a newer one. Among the 20 highest-ranked models on the capability index, these have no score yet:
- Claude Fable 5.2
-
Gemini 4 Carbon - Claude Opus 5.5
- GPT-6.1 Sol
- and 14 more
- 1 GPT-6 Astra ultra; original evaluation inputs 0.71
- 2 Claude Fable 5.1 max; original evaluation inputs 0.62
- 3 Gemini 3.8 Flash high; original evaluation inputs 0.37
- 4 DeepSeek V4 Flash Vision max; original evaluation inputs 0.34
- 5 Grok 4.6 xhigh; original evaluation inputs 0.33
- 6
Muse Spark 1.3 xhigh; original evaluation inputs 0.27 - 7 GLM-5.3-Flash max; original evaluation inputs 0.08
Source: Official FrontierCAD Assemblies leaderboard
| # | Entry | Score | Setting | Date | Source |
|---|---|---|---|---|---|
| 1 | GPT-6 Astra | 0.71 | ultra; original evaluation inputs, codex | Sep 2026 | Official leaderboard |
| 2 | Claude Fable 5.1 | 0.62 | max; original evaluation inputs, claude-code | Sep 2026 | Official leaderboard |
| 3 | Gemini 3.8 Flash | 0.37 | high; original evaluation inputs, gemini-cli | Sep 2026 | Official leaderboard |
| 4 | DeepSeek V4 Flash Vision | 0.34 | max; original evaluation inputs, opencode | Official leaderboard | |
| 5 | Grok 4.6 | 0.33 | xhigh; original evaluation inputs, grok-build | Aug 2026 | Official leaderboard |
| 6 | Muse Spark 1.3 | 0.27 | xhigh; original evaluation inputs, muse-code | Sep 2026 | Official leaderboard |
| 7 | GLM-5.3-Flash | 0.08 | max; original evaluation inputs, opencode | Aug 2026 | Official leaderboard |
Number one over time
Lab claims count from the day the card came out, independent runs from the model’s release.
Source: Official FrontierCAD Assemblies leaderboard
Scores over time
All the scores we have, one dot each, so a model tested at several settings appears more than once.
Source: Official FrontierCAD Assemblies leaderboard
No lab model card reports it yet
We read the headline results of 522 launch posts from 52 labs, and none of them list this benchmark.
More in Engineering and CAD
- EEBench In use 13 electrical-engineering circuit and PCB design tasks graded by deterministic build checks, simulation and BOM constraints. No LLM judge. Gemini 4 Carbon 83.9%
- Parametric CAD Bench In use Agents author editable parametric FreeCAD part designs, graded against held-back references for geometry and spec consistency. Claude Opus 5.5 61%
- 3DCodeBench In use Agentic procedural 3D modeling via code: author engine-ready 3D assets through software APIs. Graded for executability and shape fidelity.
LongCat-2.5 55.4% - CADGenBench Only in system cards Agents construct CAD models from design prompts. The generation split is graded for executability and geometric correctness. Claude Fable 5 45.1%
- PartBench Only in system cards Parametric mechanical part design and CAD-style generation, graded for geometric and manufacturing fidelity. Grok 4.6 58.5%
- LMBuild Not in model cards Language model agents design 3D structures as parts, joints, materials and assembly steps. Scoring checks that the structures can be built and work as intended. GPT-6 Astra 66.5%