selfhost.directory

The index of self-hostable projects — ranked, categorized and served with live updates.
Own your data and stop giving it to big tech.

ATH-MaaS logo

OvisOCR3

ATH-MaaS·Instruct·released 21 Oct 2026

Apache 2.0853M256K ctx ≈2.4 GB
Runs on CPU or any GPU≈1.6 GB of weights
ollama run hf.co/mradermacher/OvisOCR3-GGUF:Q4_K_M
All 6 runners ↓

OvisOCR3 is a unified 0.8B multimodal model for OCR and document understanding. It integrates document parsing, general OCR, key information extraction, document figure captioning, and cross-page table reconstruction into a single compact model.

How much VRAM does OvisOCR3 need?

At FP16 / BF16 OvisOCR3 needs about 2.4 GB of GPU memory.

--vram
any
--gpus
--quant
QuantisationWeightsTotal
4-bitRecommended default for local use0.5 GB file≈1.1 GB
8-bitLossless in practice, twice the size of Q40.8 GB file≈1.4 GB
FP16 / BF16recommendedFull precision, as released1.4 GB file≈2.2 GB

Estimates for the weights plus runtime overhead; activations vary with resolution and batch size. Pick a GPU above and the fit column checks every row against it. Weights marked “file” use the real GGUF download size; “est.” rows are computed from the parameter count.

Can the RTX 4060 run OvisOCR3? →

Run OvisOCR3 locally

--quantQ4_K_Mdefault; pick your GPU to tailor itmradermacher/OvisOCR3-GGUF

Q4_K_M straight from Hugging Facedocs
ollama run hf.co/mradermacher/OvisOCR3-GGUF:Q4_K_M

Install Ollama

OpenAI-compatible server with a web UIdocs
llama-server -hf mradermacher/OvisOCR3-GGUF:Q4_K_M

Downloads the GGUF on first run. Installed through llama.app? The same flags work with `llama serve`.

Install llama.cpp

Download with the lms CLIdocs
lms get mradermacher/OvisOCR3-GGUF@Q4_K_M

Then load it from the LM Studio app or with `lms load`.

Install LM Studio

Serve an OpenAI-compatible APIdocs
vllm serve ATH-MaaS/OvisOCR3
Smaller AWQ builddocs
vllm serve leleroyn/OvisOCR3-AWQ-W4A16
Docker Compose (NVIDIA)docs
yaml
services:
  vllm:
    image: vllm/vllm-openai:latest
    command: ATH-MaaS/OvisOCR3
    ports:
      - "8000:8000"
    ipc: host
    volumes:
      - ~/.cache/huggingface:/root/.cache/huggingface
    deploy:
      resources:
        reservations:
          devices:
            - driver: nvidia
              count: all
              capabilities: [gpu]

API on port 8000.

Install vLLM

Serve an OpenAI-compatible APIdocs
sglang serve ATH-MaaS/OvisOCR3 --host 0.0.0.0 --port 30000
Docker Compose (NVIDIA)docs
yaml
services:
  sglang:
    image: lmsysorg/sglang:latest
    command: sglang serve ATH-MaaS/OvisOCR3 --host 0.0.0.0 --port 30000
    ports:
      - "30000:30000"
    ipc: host
    volumes:
      - ~/.cache/huggingface:/root/.cache/huggingface
    deploy:
      resources:
        reservations:
          devices:
            - driver: nvidia
              count: all
              capabilities: [gpu]

API on port 30000.

Install SGLang

Ask about an imagedocs
python
from transformers import pipeline

pipe = pipeline("image-text-to-text", model="ATH-MaaS/OvisOCR3", dtype="auto", device_map="auto")
messages = [
    {
        "role": "user",
        "content": [
            {"type": "image", "url": "https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/bee.jpg"},
            {"type": "text", "text": "Describe this image."},
        ],
    }
]
out = pipe(text=messages, max_new_tokens=256)
print(out[0]["generated_text"][-1]["content"])

Install Transformers

Benchmarks

BenchmarkScoreRankSourceDate
OmniDocBench (Real5)93.7%#1 / 22 independentNov 2026

Self-reported scores come from the model card and are not independently verified. Rank is among all models with a score on that benchmark.

Quantised versions and downloads

RepositoryFormatQuantSizeDownloads
mradermacher/OvisOCR3-GGUFGGUFQ2_K0.4 GB18.9K
mradermacher/OvisOCR3-GGUFGGUFQ3_K_S0.4 GB18.9K
mradermacher/OvisOCR3-GGUFGGUFQ3_K_M0.5 GB18.9K
mradermacher/OvisOCR3-GGUFGGUFQ3_K_L0.5 GB18.9K
mradermacher/OvisOCR3-GGUFGGUFIQ4_XS0.5 GB18.9K
mradermacher/OvisOCR3-GGUFGGUFQ4_K_S0.5 GB18.9K
mradermacher/OvisOCR3-GGUFGGUFQ4_K_M0.5 GB18.9K
mradermacher/OvisOCR3-GGUFGGUFQ5_K_S0.6 GB18.9K
mradermacher/OvisOCR3-GGUFGGUFQ5_K_M0.6 GB18.9K
mradermacher/OvisOCR3-GGUFGGUFQ6_K0.6 GB18.9K

About OvisOCR3

OvisOCR3 is a unified 0.8B multimodal model for OCR and document understanding. It integrates document parsing, general OCR, key information extraction, document figure captioning, and cross-page table reconstruction into a single compact model. These capabilities are selected through task-specific prompts, bringing recognition, extraction, and understanding together without requiring separate model deployments for each task. Built on Qwen3.5-0.8B, OvisOCR3 is trained in four stages: continued pretraining, mid-training, on-policy distillation, and reinforcement learning. This training process builds a strong foundation in text recognition and document structure understanding, then refines these capabilities through domain-specific supervision and rewards. OvisOCR3 achieves state-of-the-art performance among models of comparable size across all five domains, demonstrating that a single compact model can combine strong document parsing with a broad range of OCR and document-understanding capabilities. The task prompts specify the output contracts. The examples below use these files directly. Performance comparisons across five task domains are reported below. -- denotes an unreported value, not zero. Bold scores indicate the best result in each column.

Full model card on Hugging Face →

Alternatives to OvisOCR3

All alternatives →
datalab-to logo

surya-ocr-2

datalab-to·OCR

≈2.0 GB

It works on a range of documents (see usage and benchmarks).

686M256K ctx OpenRAIL
668.5KMay 2026gguf
lightonai logo

LightOnOCR-2-1B

lightonai·OCR

≈2.8 GB

Best OCR model . LightOnOCR-2-1B is LightOn's flagship OCR model, refined with RLVR training for maximum accuracy. We recommend this variant for most OCR tasks.

1B16K ctx Apache 2.0
726.3KJan 2026gguf
datalab-to logo

surya-ocr-2.5

datalab-to·OCR

≈2.0 GB

We also ship smaller models for line-level text detection and ocr error detection. It works on a range of documents (see usage and benchmarks).

689M256K ctx OpenRAIL
129.8KOct 2026gguf
lightonai logo

LightOnOCR-3-0.8B

lightonai·OCR

≈2.4 GB

Smallest model. LightOnOCR-3-0.8B is the lightest model of LightOn's LightOnOCR-3 family, with the same layout and visual understanding features.

853M256K ctx Apache 2.0
15.1KOct 2026gguf
ATH-MaaS logo

OvisOCR2

ATH-MaaS·OCR

≈2.4 GB

We are pleased to announce the release of OvisOCR2, a compact 0.8B end-to-end model for page-level document parsing. Given a document page image, OvisOCR2 generates a Markdown representation in natural reading order, covering text, formulas, tables, and visual regions.

853M256K ctx Apache 2.0
25KJul 2026gguf
PaddlePaddle logo

PaddleOCR-VL-1.6

PaddlePaddle·OCR

≈2.6 GB

PaddleOCR-VL-1.6: Expanding the Frontier of Document Parsing with Under-Optimized Region Refinement and Progressive Post-Training

959M128K ctx Apache 2.0
99.7KMay 2026gguf

OvisOCR3 FAQ

How much VRAM does OvisOCR3 need?
About 2.4 GB at FP16 / BF16.
Can OvisOCR3 run on the RTX 4060?
Yes. It fits at FP16 / BF16 (about 2.2 GB). It fits comfortably on the Apple M4 (16 GB), Radeon RX 6700 XT, RTX 3060 12GB.
Can I use OvisOCR3 commercially?
Yes. It is released under Apache 2.0. Permissive: use, modify and sell freely; keep the notice and licence text.
How do I run OvisOCR3 with Ollama or llama.cpp?
With Ollama: ollama run hf.co/mradermacher/OvisOCR3-GGUF:Q4_K_M. With llama.cpp: llama-server -hf mradermacher/OvisOCR3-GGUF:Q4_K_M, which starts an OpenAI-compatible server.

Data from Hugging Face, the model card and public leaderboards. Memory and speed figures are estimates; real numbers depend on the runtime, context length and drivers. Always check the licence yourself before commercial use.