Constarium
← Search

Text · dataset · 2026

GPT-6 FEniCSx Single-Attempt Pass Rate and Solver Corpus

Listed in Hugging Face Datasets

GPT-6 FEniCSx Single-Attempt Pass Rate and Solver Corpus A community evaluation and reusable scientific-code corpus by Haoming Luo.

Description

A collection of 496 original, unedited first-attempt Python/FEniCSx solvers generated by GPT-6 Astra through the Codex CLI and passing the original PDEAgent-Bench accuracy and runtime gates in our evaluation environment. Of 645 requested benchmark tasks, 644 have been evaluated: 496/644 = 77.02% first-attempt success.

One task remains blocked by a… See the full description on the dataset page: huggingface.co/datasets/HaomingLuo/GPT-6-FEniCSx-Single-Attempt-Pass-Rate.

Links

Topics

Stated by source
tabular · text
Inferred from text
Theory of computation 71%
Provenance · 1 source records, 11 field assertions
SourceKeyLast seenRaw
Hugging Face DatasetsHaomingLuo/GPT-6-FEniCSx-Single-Attempt-Pass-Rate12 d agoJSON v1
FieldAssertionExtractorEvidence
access_levelsource · Hugging Faceconnector:huggingface@1.0.0/gated
concepts[field].anzsrc:group:4613enrichment · Hugging Facetaxonomy-embedding@1.1.0title+keywords+description (71%)
concepts[field].local:field:computer-science-aimapping · Hugging Faceconnector:huggingface@1.0.0
concepts[modality].hf_modality:tabularsource · Hugging Faceconnector:huggingface@1.0.0
concepts[modality].hf_modality:textsource · Hugging Faceconnector:huggingface@1.0.0
created_datesource · Hugging Faceconnector:huggingface@1.0.0
descriptionsource · Hugging Faceconnector:huggingface@1.0.0/description
licensesource · Hugging Faceconnector:huggingface@1.0.0/tags[license:*]
publication_datesource · Hugging Faceconnector:huggingface@1.0.0
titlesource · Hugging Faceconnector:huggingface@1.0.0/id
updated_datesource · Hugging Faceconnector:huggingface@1.0.0