RE

Realocr

RealocrGGUF by mradermacherUpdated 2026-10-17
Download on Hugging Face
Parameters1.88B
Context256K
Layers25
Architectureqwen35
Smallest quant1.0 GB
Quick answer

Realocr needs about 2.2 GB of memory at Q4_K_M (a 1.3 GB file) with an 8K context window, so it fits on a 4 GB GPU. This repository has 12 quants, from 1.0 GB (Q2_K) to 3.8 GB (F16). Longer context needs more: at 32K tokens the KV cache alone adds about 1.7 GB.

Realocr – 1.88B Qwen3.5 model for document parsing

Realocr is a 1.88B‑parameter Qwen3.5 model designed for end‑to‑end document parsing. It accepts a page image and outputs full‑page Markdown. The model supports reasoning, tools, and vision.

Good at

  • Document parsing
  • Vision‑enabled text extraction
  • Reasoning over image content

Limits

  • No vision support on older GPUs due to Conv3d kernel
  • Requires 8k context for full output
  • License is apache‑2.0

Best for

  • Generating Markdown from scanned pages
  • Automating MDPBench submissions
  • Batch processing of document images

Who it’s for. Developers building document‑to‑text pipelines or researchers evaluating vision‑language models. The model is suitable for users who need accurate Markdown output from images and can run it on compatible hardware.

Quants & memory

QuantFileEst. memoryvs VRAMResultDownload
Q2_K1.0 GB 1.9 GB
Fits Get file →
Q3_K_S1.0 GB 1.9 GB
Fits Get file →
Q3_K_M1.1 GB 2.0 GB
Fits Get file →
Q3_K_L1.2 GB 2.1 GB
Fits Get file →
IQ4_XS1.2 GB 2.1 GB
Fits Get file →
Q4_K_S1.2 GB 2.1 GB
Fits Get file →
Q4_K_M1.3 GB 2.2 GB
Fits Get file →
Q5_K_S1.4 GB 2.3 GB
Fits Get file →
Q5_K_M1.4 GB 2.3 GB
Fits Get file →
Q6_K1.6 GB 2.5 GB
Fits Get file →
Q8_02.0 GB 2.9 GB
Fits Get file →
F163.8 GB 4.7 GB
Fits Get file →

For 8 GB VRAM use Q4_K_M (1.27 GB, ~2.2 GB memory at 8k context). For 12 GB VRAM use Q5_K_M (1.41 GB, ~2.3 GB memory). For higher precision, use F16 (3.78 GB, ~4.7 GB memory).

What GPU do you need for Realocr?

These figures add the file size, an fp16 KV cache at an 8K context window (worked out from this model’s own header) and about 0.5 GB for the runtime. They are estimates, so leave a little headroom, and expect to use less context or a smaller quant if you are right at the limit.

  • 4 GB VRAM: the best quant that fits is Q8_0 (2.0 GB file, about 2.9 GB in use).
  • 6 GB VRAM: the best quant that fits is F16 (3.8 GB file, about 4.7 GB in use).

Which Realocr quant should you choose?

Q4_K_M is the usual starting point for Realocr: it balances file size and quality well. Here it is a 1.3 GB file.

The Q8_0 file is 2.0 GB, about 1.6 times the size of Q4_K_M. It keeps the weights closer to the original, which only matters if you have the memory to spare.

If memory is tight, the smallest file is Q2_K at 1.0 GB, with the quality trade-offs that come with fewer bits. Use it only when memory is very tight. Expect a clear drop in quality, which is usually worse on small models and easier to live with on large ones.

New to quant names? Read GGUF quantization explained.

Specifications

Architecture from the GGUF header

Architectureqwen35
Parameters1.88B
Layers25
Trained context262,144 tokens
KV cache (fp16)50 KB per token

Source

Quantized bymradermacher
FamilyRealocr
Licenseapache-2.0
Downloads (HF)709
Last updated2026-10-17

Frequently asked questions about Realocr

How much VRAM does Realocr need?

About 2.2 GB at Q4_K_M with an 8K context window. That is the 1.3 GB file, plus the KV cache and roughly 0.5 GB of runtime overhead. Smaller quants need less and longer context needs more.

Can I run Realocr on an 8 GB GPU?

Yes, with F16: the 3.8 GB file needs about 4.7 GB at 8K context. Larger quants in this repository do not fit in 8 GB.

What is the file size of Realocr GGUF?

It depends on the quant: Q2_K is 1.0 GB, Q3_K_S is 1.0 GB, Q3_K_M is 1.1 GB, Q3_K_L is 1.2 GB, IQ4_XS is 1.2 GB, Q4_K_S is 1.2 GB and 6 more.

What context length does Realocr support?

The GGUF header lists a trained context length of 262,144 tokens (256K). The KV cache grows with the context you use: about 0.4 GB at 8K and 1.7 GB at 32K.

Where can I download Realocr GGUF?

The files are on Hugging Face in the mradermacher/realocr-GGUF repository. The download buttons on this page open the file on Hugging Face. ggufmodels.com does not host any model files.

What license does Realocr use?

Hugging Face lists the license as apache-2.0. Licenses can differ between the quantized files and the original model, so check the repository and the base model before commercial use.

Can I run Realocr on 8 GB VRAM?

Yes, the Q4_K_M quant uses about 2.2 GB memory at 8k context, fitting within 8 GB VRAM.

Does Realocr support vision on all GPUs?

Vision patch‑embedding uses Conv3d, which may hit a cuDNN kernel issue on some GPUs; the provided script replaces it with a Linear matmul.

What license applies to Realocr?

The model is released under the apache‑2.0 license.

Similar models

Browse all →

More from mradermacher

All by mradermacher →

Other tiny models

All tiny models →