cstr/PP-OCRv6-medium-rec-GGUF

PP-OCRv6 medium recognizer — CrispEmbed GGUF

Files: PP-OCRv6_medium_rec-f16.gguf and PP-OCRv6_medium_rec-crispasr-q4_k-policy.gguf.

The policy-q4 container intentionally keeps the complete PP-OCRv6 recognizer graph in F16: quantizing intermediate CNN/SVTR weights caused compounding CTC drift. Source: PaddlePaddle PP-OCRv6, Apache-2.0.

Parity on tests/regression/images/fox.png: F16 and policy-q4 both pass the 0.999 cosine floor across the full native diff path; worst stage cosine is 0.999543 and logits cosine is 0.999543.

Canonical CrispEmbed publisher: cstr. Large source and working files are kept on /Volumes/backups/ai/crispembed-gguf/.

Provenance and EU AI Act Art. 53 note

• Upstream model: PaddlePaddle PP-OCRv6.
• Upstream licence: apache-2.0. This repository redistributes under the same terms; it grants no rights the upstream licence does not.
• What was done here: format conversion and/or quantisation only (GGUF). No training, no fine-tuning, no merging, no distillation, no change to architecture, vocabulary or capability. Only the numeric representation of the upstream weights differs.
• Training data: documented — where it is documented at all — by the upstream provider; see the upstream model card. No training data was used, added or selected by this repository.
• Provider status: under Regulation (EU) 2024/1689 the upstream authors remain the provider of this model. Converting the serialisation format does not make this repository the provider of a new general-purpose AI model, and no such claim is made. Questions about training content, copyright policy or model capability belong upstream.