OL

olmOCR 7B Thai v3

OLMoGGUF by mradermacherUpdated 2025-07-10
Download on Hugging Face
Parameters7.62B
Context32K
Layers28
Architectureqwen2vl
Smallest quant1.9 GB
Quick answer

olmOCR 7B Thai v3 needs about 5.7 GB of memory at Q4_K_M (a 4.7 GB file) with an 8K context window, so it fits on a 6 GB GPU. This repository has 24 quants, from 1.9 GB (IQ1_S) to 6.3 GB (Q6_K). Longer context needs more: at 32K tokens the KV cache alone adds about 1.9 GB.

olmOCR 7B Thai v3 – a vision‑language OCR model fine‑tuned for Thai

olmOCR 7B Thai v3 is a 7.62 B parameter vision‑language model based on Qwen2‑VL‑7B‑Instruct. It is fine‑tuned on 250K Thai documents for high‑accuracy OCR of PDFs and receipts. The model is released under the MIT license and supports command‑line and API inference.

Good at

  • Thai OCR from PDFs
  • High‑accuracy table extraction
  • Command‑line and API usage

Limits

  • No vision support in the base repo
  • Requires 8 k context for full memory estimates
  • Large VRAM needed for higher‑quality quants

Best for

  • Document digitization in Thai
  • Receipt processing
  • Table extraction from PDFs

Who it’s for. Developers building OCR pipelines for Thai language documents. Researchers needing a fine‑tuned vision‑language model for Thai text extraction. Teams looking for an open‑source, MIT‑licensed OCR solution.

Quants & memory

QuantFileEst. memoryvs VRAMResultDownload
IQ1_S1.9 GB 2.9 GB
Fits Get file →
IQ1_M2.0 GB 3.0 GB
Fits Get file →
IQ2_XXS2.3 GB 3.2 GB
Fits Get file →
IQ2_XS2.5 GB 3.4 GB
Fits Get file →
IQ2_S2.6 GB 3.6 GB
Fits Get file →
IQ2_M2.8 GB 3.8 GB
Fits Get file →
Q2_K_S2.8 GB 3.8 GB
Fits Get file →
Q2_K3.0 GB 4.0 GB
Fits Get file →
IQ3_XXS3.1 GB 4.1 GB
Fits Get file →
IQ3_XS3.3 GB 4.3 GB
Fits Get file →
Q3_K_S3.5 GB 4.5 GB
Fits Get file →
IQ3_S3.5 GB 4.5 GB
Fits Get file →
IQ3_M3.6 GB 4.5 GB
Fits Get file →
Q3_K_M3.8 GB 4.8 GB
Fits Get file →
Q3_K_L4.1 GB 5.1 GB
Fits Get file →
IQ4_XS4.2 GB 5.2 GB
Fits Get file →
IQ4_NL4.4 GB 5.4 GB
Fits Get file →
Q4_04.4 GB 5.4 GB
Fits Get file →
Q4_K_S4.5 GB 5.4 GB
Fits Get file →
Q4_K_M4.7 GB 5.7 GB
Fits Get file →
Q4_14.9 GB 5.8 GB
Fits Get file →
Q5_K_S5.3 GB 6.3 GB
Fits Get file →
Q5_K_M5.4 GB 6.4 GB
Fits Get file →
Q6_K6.3 GB 7.2 GB
Tight Get file →

For 8 GB VRAM, use IQ1_S (1.9 GB) or IQ1_M (2.04 GB). If 12 GB VRAM is available, Q4_K_M (4.68 GB) offers higher quality. For 16 GB VRAM, Q5_K_M (5.44 GB) or Q6_K (6.25 GB) provide better accuracy. Memory estimates are approximate.

What GPU do you need for olmOCR 7B Thai v3?

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 Q2_K (3.0 GB file, about 4.0 GB in use).
  • 6 GB VRAM: the best quant that fits is Q4_1 (4.9 GB file, about 5.8 GB in use).
  • 8 GB VRAM: the best quant that fits is Q6_K (6.3 GB file, about 7.2 GB in use).

Which olmOCR 7B Thai v3 quant should you choose?

Q4_K_M is the usual starting point for olmOCR 7B Thai v3: it balances file size and quality well. Here it is a 4.7 GB file.

If memory is tight, the smallest file is IQ1_S at 1.9 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

Architectureqwen2vl
Parameters7.62B
Layers28
Trained context32,768 tokens
KV cache (fp16)56 KB per token

Source

Quantized bymradermacher
FamilyOLMo
Licensemit
Downloads (HF)510
Last updated2025-07-10

Frequently asked questions about olmOCR 7B Thai v3

How much VRAM does olmOCR 7B Thai v3 need?

About 5.7 GB at Q4_K_M with an 8K context window. That is the 4.7 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 olmOCR 7B Thai v3 on an 8 GB GPU?

Yes, with Q6_K: the 6.3 GB file needs about 7.2 GB at 8K context. Larger quants in this repository do not fit in 8 GB.

What is the file size of olmOCR 7B Thai v3 GGUF?

It depends on the quant: IQ1_S is 1.9 GB, IQ1_M is 2.0 GB, IQ2_XXS is 2.3 GB, IQ2_XS is 2.5 GB, IQ2_S is 2.6 GB, IQ2_M is 2.8 GB and 18 more.

What context length does olmOCR 7B Thai v3 support?

The GGUF header lists a trained context length of 32,768 tokens (32K). The KV cache grows with the context you use: about 0.5 GB at 8K and 1.9 GB at 32K.

Where can I download olmOCR 7B Thai v3 GGUF?

The files are on Hugging Face in the mradermacher/olmOCR-7B-thai-v3-i1-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 olmOCR 7B Thai v3 use?

Hugging Face lists the license as mit. 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 this model on 8 GB VRAM?

Yes, use IQ1_S or IQ1_M quant files, which require about 2.9 GB or 3 GB of VRAM at 8 k context.

Does this model support other languages besides Thai?

The fine‑tuned model focuses on Thai; other languages are not guaranteed to perform well.

Is there a vision model file?

Yes, the repository contains mmproj files for vision support, available in the static GGUF repository.

Other GGUFs of olmOCR-7B-thai-v3

Similar models

Browse all →

More from mradermacher

All by mradermacher →

More OLMo models

All OLMo →

Other small models

All small models →