smt-fp-grandstaff-GGUF

SMT++ Full-Page Grandstaff — GGUF

GGUF conversions of PRAIG/smt-fp-grandstaff for CrispEmbed's dependency-free smt_ocr inference engine (pure C/C++ + ggml, no Python/PyTorch at runtime).

Full-page optical music recognition: a whole pianoform sheet-music page → bekern token sequence, in one end-to-end pass (no staff/system segmentation).

Files

q4_k is not provided — even with the encoder + LM head kept at Q8_0 it only reaches ~0.5% token-CER while ending up larger than q5_k (the ConvNext encoder is most of the weights), and plain q4_k degenerates into repetition. Use q5_k.

Architecture

ConvNext encoder (3 stages [64,128,256], 16× reduction) + 8-layer cross-attention Transformer decoder (d_model 256, 4 heads, 1× ReLU FFN), sinusoidal 1-D/2-D positional encodings, greedy AR decode to a 181-token bekern vocabulary (maxlen 4353). This is the antoniorv6/SMT forward: scaled dot-product attention (d_head**-0.5), no ReLU before the linear LM head; input is used at native resolution (reduce_ratio=1.0) and grayscaled (no inversion — matching SMT/data_augmentation.convert_img_to_tensor).

Usage

crispembed -m smt-fp-grandstaff-q8_0.gguf --image page.png
# or let the model manager fetch it:
crispembed --model smt-fp --image page.png

License & attribution

MIT, inherited from the upstream weights (PRAIG/smt-fp-grandstaff) and code (antoniorv6/SMT).

If you use this model, please cite the paper:

A. Ríos-Vila, D. Rizo, J. M. Iñesta, J. Calvo-Zaragoza. End-to-End Full-Page Optical Music Recognition for Pianoform Sheet Music. International Journal of Computer Vision (IJCV), 2026. arXiv:2405.12105.

Conversions produced with models/convert-smt-to-gguf.py from CrispEmbed.

Provenance and EU AI Act Art. 53 note

• Upstream model: PRAIG/smt-fp-grandstaff — published by PRAIG.
• Upstream licence: mit. 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. No training-content summary was found on the upstream model card at the time of writing; that documentation gap is upstream's and is not filled here.
• 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.