Granite Vision Crispembed GGUF

--- license: apache-2.0 tags: [gguf, ocr, vlm, granite, siglip, document-understanding, crispembed] base_model: ibm-granite/granite-vision-3.3-2b library_name: crispembed pipeline_tag: image-to-text --- # Granite Vision 3.3-2B — CrispEmbed GGUF GGUF conversion of [ibm-granite/granite-vision-3.3-2b](https://huggingface.co/ibm-granite/granite-vision-3.3-2b) for [CrispEmbed](https://github.com/CrispStrobe/CrispEmbed). OCRBench 852 — highest score for models ≤3B parameters. ## Architecture - **Vision**: SigLIP ViT-SO400M (27 layers, 1152d, 384px, multi-layer features [-24,-20,-12,-1]) - **Projector**: 2-layer MLP (4×1152 → 2048) - **LLM**: Granite-3.1-2B (40 layers, 2048d, GQA 32/8) - `embedding_multiplier=12.0`, `residual_multiplier=0.22`, `logits_scaling=8.0` ## Models | File | Quant | Size | Notes | |------|-------|------|-------| | `granite-vision-3.3-2b-f16.gguf` | F16 | 5.6 GB | Full precision | | `granite-vision-3.3-2b-q8_0.gguf` | Q8_0 | 3.2 GB | Best quality/size | | `granite-vision-3.3-2b-q4_k.gguf` | Q4_K | 1.9 GB | Smallest (vision stays F16) | | `granite-vision-ref.gguf` | F32 | 36 MB | Reference activations for parity | ## Parity Vision encoder + projector verified via crispembed-diff: - `vis_features_concat`: cos_min=0.999990 - `projector`: cos_min=0.999972 ## Usage ```bash crispembed --ocr granite-vision-3.3-2b-q8_0.gguf document.png ``` ## License Apache-2.0 ([ibm-granite/granite-vision-3.3-2b](https://huggingface.co/ibm-granite/granite-vision-3.3-2b)) ## Provenance and EU AI Act Art. 53 note - **Upstream model:** [ibm-granite/granite-vision-3.3-2b](https://huggingface.co/ibm-granite/granite-vision-3.3-2b) — published by `ibm-granite`. - **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 th...

Source context: 263 downloads · 1 likes · Pipeline image-to-text · Library crispembed · Repo cstr/granite-vision-crispembed-GGUF

Source context: 721 downloads · 1 likes · Pipeline image-to-text · Library crispembed · Repo cstr/granite-vision-crispembed-GGUF