cstr/qari-ocr-crispembed-GGUF overview
Qari OCR CrispEmbed GGUF GGUF conversion of NAMAA Space/Qari OCR 0.2.2.1 VL 2B Instruct https://huggingface.co/NAMAA Space/Qari OCR 0.2.2.1 VL 2B Instruct Apac…
Runs locally from ~5.3 MB disk (4 GB VRAM class GPUs with llama.cpp / guIDE).
Repository Files & Downloads
Model Details
| Model ID | cstr/qari-ocr-crispembed-GGUF |
|---|---|
| Author | cstr |
| Pipeline | image-text-to-text |
| License | apache-2.0 |
| Base model | NAMAA-Space/Qari-OCR-0.2.2.1-VL-2B-Instruct |
| Last modified | 2026-06-21T06:23:08.000Z |
Model README
---
license: apache-2.0
tags:
- ocr
- arabic
- gguf
- crispembed
- qwen2-vl
base_model: NAMAA-Space/Qari-OCR-0.2.2.1-VL-2B-Instruct
language: ar
pipeline_tag: image-text-to-text
---
Qari-OCR CrispEmbed GGUF
GGUF conversion of NAMAA-Space/Qari-OCR-0.2.2.1-VL-2B-Instruct (Apache-2.0) for use with CrispEmbed.
Model
Arabic OCR with full diacritics (tashkeel) support. Fine-tuned from Qwen2-VL-2B-Instruct via LoRA (r=16, alpha=16, 324 adapter pairs) on 50K Arabic OCR samples.
- Architecture: Qwen2-VL-2B (32L ViT 1280d + spatial merger + 28L Qwen2 LLM 1536d, GQA 12/2)
- Parameters: 2B
- Performance: WER=0.221, CER=0.059, BLEU=0.597
- Training: 50K Arabic OCR records, 1 epoch, LoRA on attention+MLP
Files
| File | Type | Size |
|------|------|------|
| qari-ocr-2b-f16.gguf | F16 | 4.7 GB |
| qari-ocr-2b-q8_0.gguf | Q8_0 | 2.3 GB |
| qari-ocr-2b-q4_k.gguf | Q4_K | 1.6 GB |
Usage
Uses the same qwen2vl engine as other Qwen2-VL models in CrispEmbed.
Conversion
LoRA adapter merged into full-precision Qwen2-VL-2B-Instruct base weights (324 pairs, tensor-by-tensor), then converted to GGUF via CrispEmbed converter. Quantized with crispembed-quantize (vision weights at Q8_0 floor).
License
Apache-2.0 (NAMAA-Space/Qari-OCR).
Run cstr/qari-ocr-crispembed-GGUF with guIDE
Download guIDE — the AI-native code editor with local LLM inference and 69 built-in tools.
Source: Hugging Face · Compare models