BlogLLM Benchmark Scores as Published, Side by Side

BenchCAD: Results as Published

Every result the catalog holds for BenchCAD: 17 models from 4 developers, read from 7 pages. Each number is as printed by the company named beside it; nothing is ranked or rescored. Models are listed by developer, newest first.

Area Engineering and DesignMaintainer Authors of the May 2026 arXiv paperIntroduced 2026Data version 2026103001, as of 2026-10-30

BenchCAD is a benchmark of 17,900 industrial CAD programs written in CadQuery, each checked by running it, across 106 part families such as gears, springs and drill bits. Models answer questions about drawings and code, turn an image into code, and edit code according to an instruction. The authors report that current models often capture the rough outer shape of a part but struggle to build faithful parametric programs.

BenchCAD

ModelValuePrinted bySetup as printed
GPT-6 AstraOpenAI, released 2026-09-0395.9%OpenAI, GPT-6 Astrachecked 2026-10-14; printed on 1 more page-
GPT-5.6 LunaOpenAI, released 2026-07-0963.1%OpenAI, GPT-5.6: Frontier intelligence that scales with your ambitionchecked 2026-10-15-
GPT-5.6 SolOpenAI, released 2026-07-0970.6%OpenAI, GPT-5.6: Frontier intelligence that scales with your ambitionchecked 2026-10-15; printed on 1 more page-
GPT-5.6 TerraOpenAI, released 2026-07-0962.3%OpenAI, GPT-5.6: Frontier intelligence that scales with your ambitionchecked 2026-10-15-
GPT-5.5OpenAI, released 2026-04-2344.4%OpenAI, GPT-5.6: Frontier intelligence that scales with your ambitionchecked 2026-10-15-
Claude Opus 5.5Anthropic, released 2026-09-2296.2%Moonshot AI, Introducing Kimi K3.1 Prochecked 2026-10-24Effort: max; Tools: with tools; Fallback: with fallback
Claude Fable 5.1Anthropic, released 2026-09-0192.6%Moonshot AI, Introducing Kimi K3.1 Prochecked 2026-10-24; printed on 1 more pageEffort: max; Tools: with tools; Fallback: with fallback
Claude Opus 5Anthropic, released 2026-07-2482.1%OpenAI, GPT-6 Astrachecked 2026-10-14-
Claude Fable 5Anthropic, released 2026-06-0967.5%OpenAI, GPT-6 Astrachecked 2026-10-14-
Claude Mythos 5Anthropic, released 2026-06-0938.4%OpenAI, GPT-5.6: Frontier intelligence that scales with your ambitionchecked 2026-10-15-
Claude Opus 4.8Anthropic, released 2026-05-2827.3%OpenAI, GPT-5.6: Frontier intelligence that scales with your ambitionchecked 2026-10-15-
Claude Mythos PreviewAnthropic, released 2026-04-0735.5%OpenAI, GPT-5.6: Frontier intelligence that scales with your ambitionchecked 2026-10-15-
DeepSeek-V4.1-Flash-1017DeepSeek, released 2026-10-1781.6%Moonshot AI, Introducing Kimi K3.1 Prochecked 2026-10-24Effort: max; Harness: Claude Code; Tools: with tools
Kimi K3.1 ProMoonshot AI, released 2026-10-2392.7%Moonshot AI, Introducing Kimi K3.1 Prochecked 2026-10-24Effort: max; Harness: Claude Code; Tools: with tools

BenchCAD (python tool)

ModelValuePrinted bySetup as printed
GPT-5.6 LunaOpenAI, released 2026-07-0973.9%OpenAI, GPT-5.6: Frontier intelligence that scales with your ambitionchecked 2026-10-15Tools: python tool
GPT-5.6 SolOpenAI, released 2026-07-0983.4%OpenAI, GPT-5.6: Frontier intelligence that scales with your ambitionchecked 2026-10-15Tools: python tool
GPT-5.6 TerraOpenAI, released 2026-07-0978.2%OpenAI, GPT-5.6: Frontier intelligence that scales with your ambitionchecked 2026-10-15Tools: python tool
GPT-5.5OpenAI, released 2026-04-2355.8%OpenAI, GPT-5.6: Frontier intelligence that scales with your ambitionchecked 2026-10-15Tools: python tool
Claude Mythos 5Anthropic, released 2026-06-0965%OpenAI, GPT-5.6: Frontier intelligence that scales with your ambitionchecked 2026-10-15Tools: python tool
Claude Opus 4.8Anthropic, released 2026-05-2851.8%OpenAI, GPT-5.6: Frontier intelligence that scales with your ambitionchecked 2026-10-15Tools: python tool
Claude Mythos PreviewAnthropic, released 2026-04-0761%OpenAI, GPT-5.6: Frontier intelligence that scales with your ambitionchecked 2026-10-15Tools: python tool

BenchCAD Vision2Code, no tools (voxel IoU)

ModelValuePrinted bySetup as printed
Claude Sonnet 5.5Anthropic, released 2026-09-280.747Anthropic, Claude Sonnet 5.5 System Cardchecked 2026-10-19Effort: max
Claude Opus 5.5Anthropic, released 2026-09-220.73Anthropic, Claude Opus 5.5 System Cardchecked 2026-10-25; printed on 1 more pageTools: none
Claude Fable 5.1Anthropic, released 2026-09-010.437Anthropic, System Card: Claude Fable 5.1 & Claude Mythos 5.1checked 2026-10-25; printed on 2 more pagesTools: none
Claude Opus 5Anthropic, released 2026-07-240.497Anthropic, Claude Opus 5.5 System Cardchecked 2026-10-25; printed on 1 more pageTools: none
Claude Fable 5Anthropic, released 2026-06-090.376Anthropic, System Card: Claude Fable 5.1 & Claude Mythos 5.1checked 2026-10-25Tools: none

BenchCAD Vision2Code, with tools (voxel IoU)

ModelValuePrinted bySetup as printed
Claude Sonnet 5.5Anthropic, released 2026-09-280.963Anthropic, Claude Sonnet 5.5 System Cardchecked 2026-10-19Effort: max; Tools: Container with the image files and standard libraries, image cropping tool
Claude Opus 5.5Anthropic, released 2026-09-220.962Anthropic, Claude Opus 5.5 System Cardchecked 2026-10-25; printed on 1 more pageTools: enabled
Claude Fable 5.1Anthropic, released 2026-09-010.843Anthropic, System Card: Claude Fable 5.1 & Claude Mythos 5.1checked 2026-10-25; printed on 2 more pagesTools: enabled
Claude Opus 5Anthropic, released 2026-07-240.899Anthropic, Claude Opus 5.5 System Cardchecked 2026-10-25; printed on 1 more pageTools: enabled
Claude Fable 5Anthropic, released 2026-06-090.675Anthropic, System Card: Claude Fable 5.1 & Claude Mythos 5.1checked 2026-10-25Tools: enabled

BenchCAD w/o tools

ModelValuePrinted bySetup as printed
Claude Opus 5.5Anthropic, released 2026-09-2273%Moonshot AI, Introducing Kimi K3.1 Prochecked 2026-10-24Effort: max; Tools: none; Fallback: with fallback
Claude Fable 5.1Anthropic, released 2026-09-0160.6%Moonshot AI, Introducing Kimi K3.1 Prochecked 2026-10-24Effort: max; Tools: none; Fallback: with fallback
DeepSeek-V4.1-Flash-1017DeepSeek, released 2026-10-1741.2%Moonshot AI, Introducing Kimi K3.1 Prochecked 2026-10-24Effort: max; Tools: none
Kimi K3.1 ProMoonshot AI, released 2026-10-2372.8%Moonshot AI, Introducing Kimi K3.1 Prochecked 2026-10-24Effort: max; Tools: none

BenchCAD Vision2Code voxel IoU

ModelValuePrinted bySetup as printed
Claude Haiku 5.5Anthropic, released 2026-10-070.67Anthropic, Claude Haiku 5.5 System Cardchecked 2026-10-19Tools: none
0.87Anthropic, Claude Haiku 5.5 System Cardchecked 2026-10-19Tools: enabled
Claude Sonnet 5.5Anthropic, released 2026-09-280.747Anthropic, Claude Haiku 5.5 System Cardchecked 2026-10-19Tools: none
0.963Anthropic, Claude Haiku 5.5 System Cardchecked 2026-10-19Tools: enabled
Claude Haiku 4.5Anthropic, released 2025-10-150.155Anthropic, Claude Haiku 5.5 System Cardchecked 2026-10-19Tools: none
0.21Anthropic, Claude Haiku 5.5 System Cardchecked 2026-10-19Tools: enabled

Every number here is read from the same catalog file as the benchmark table, data version 2026103001, and is the number the table shows for that model and test. Where one page printed a model under several setups, each value is listed. The catalog is public in armanobosyan/sugra-llm-benchmarks; a wrong number can be reported from the table, with Report an error under the value, or as an issue in that repository.