Dataset · Hugging Face

schneewolflabs/ChartDocMix-v1

Name: ChartDocMix-v1
Creator: schneewolflabs
License: apache-2.0
Keywords: task_categories:visual-question-answering, task_categories:image-text-to-text, language:en, license:apache-2.0, size_categories:100K<n<1M, format:parquet, format:optimized-parquet, modality:image, modality:text, library:datasets, library:dask, library:polars, library:mlcroissant, region:us, multimodal

ChartDocMix-v1 A focused-excellence Stage-2 corpus for A3-Doc — Schneewolf Labs' document + chart specialist VLM (Qwen3-VL ViT grafted onto the A-series / Mistral decoder via Artemis). Where ArtemisMix-v1.1 is a broad generalist mix, ChartDocMix is deliberately narrow: ~96% of the corpus is document/chart/OCR visual QA, with a small text rehearsal slice to hold identity and prevent voice drift during the full fine-tune. The target is best-in-class on the ChartQA / DocVQA /… See the full description on the dataset page: https://huggingface.co/datasets/schneewolflabs/ChartDocMix-v1.

View on Hugging FaceReport an issue

Author / organization	schneewolflabs
License (as declared)	apache-2.0
Downloads	235 (as reported by Hugging Face)
Likes	0
Created	8 June 2026
Last updated at source	8 June 2026

multimodalvlmdocument-understandingchart-understandingocrinstruction-tuningschneewolf-labsartemis

Embed this entry

<iframe src="https://silytics.si/embed/item/e4f1b55e-c500-4a63-b3fd-734b722f2e6b" width="420" height="190" style="border:0" loading="lazy" title="Silytics entry"></iframe>

Free, no tracking. More about feeds and embeds.

Popularity, last 2 days

235 now (-2 since 2026-10-16). Counts as reported by the source, snapshotted daily by Silytics.

More from schneewolflabs

Silytics shows public metadata from Hugging Face and links to the original. Information may be incomplete or out of date. Last refreshed 17 October 2026. Licenses and terms of the original apply to the work itself. Not financial, legal or professional advice.