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.

datalab-to logo

Can the Radeon RX 9070 XT run surya-ocr-2.5?

surya-ocr-2.5·689M·Radeon RX 9070 XT·16 GB VRAM

Saved in this browser, used on every model page.
Yes

Yes. The Radeon RX 9070 XT runs surya-ocr-2.5 fully in VRAM at FP16 / BF16 (about 2.0 GB of 16 GB).

best quant
FP16 / BF16
needs
2.0 GB
usable
16 GB

Estimates from the model's size.Try other contexts on the surya-ocr-2.5 page →

Every quantisation of surya-ocr-2.5 on the Radeon RX 9070 XT

QuantMemoryFit
Q4_K_M 1.0 GBfits
Q8_0 1.2 GBfits
FP16 / BF16 pick2.0 GBfits

Speed is a bandwidth-based estimate for generating text (640 GB/s); prompt processing and long contexts are slower.

Run surya-ocr-2.5 on the Radeon RX 9070 XT

Using Q8_0, the best fit for Radeon RX 9070 XT.

Ollama · Q8_0 straight from Hugging Facedocs
ollama run hf.co/datalab-to/surya-ocr-2.5-gguf:Q8_0
Ollama · Docker with ROCm (AMD)docs
docker run -d --name ollama --device /dev/kfd --device /dev/dri \
  -v ollama:/root/.ollama -p 11434:11434 ollama/ollama:rocm
docker exec -it ollama ollama run hf.co/datalab-to/surya-ocr-2.5-gguf:Q8_0

The image bundles the ROCm runtime; the host only needs the amdgpu kernel driver.

llama.cpp · OpenAI-compatible server with a web UIdocs
llama-server -hf datalab-to/surya-ocr-2.5-gguf:Q8_0

Needs a HIP (ROCm) build of llama.cpp. Downloads the GGUF on first run.

llama.cpp · Docker with ROCm (AMD)docs
docker run -d --name llama --device /dev/kfd --device /dev/dri -p 8080:8080 \
  -v ~/.cache/llama.cpp:/root/.cache/llama.cpp \
  ghcr.io/ggml-org/llama.cpp:server-rocm \
  -hf datalab-to/surya-ocr-2.5-gguf:Q8_0 --host 0.0.0.0 --port 8080 -ngl 99

Server and web UI on port 8080.

LM Studio · Download with the lms CLIdocs
lms get datalab-to/surya-ocr-2.5-gguf@Q8_0

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

vLLM · Serve an OpenAI-compatible APIdocs
vllm serve datalab-to/surya-ocr-2.5

Needs vLLM built for ROCm; it supports Instinct MI200/MI300 and Radeon RX 7900 / 9000 cards.

vLLM · Docker Compose (AMD ROCm)docs
services:
  vllm:
    image: rocm/vllm:latest
    command: vllm serve datalab-to/surya-ocr-2.5 --host 0.0.0.0 --port 8000
    ports:
      - "8000:8000"
    ipc: host
    devices:
      - /dev/kfd
      - /dev/dri
    group_add:
      - video
    security_opt:
      - seccomp=unconfined
    volumes:
      - ~/.cache/huggingface:/root/.cache/huggingface

AMD’s rocm/vllm image. API on port 8000.

SGLang · Serve an OpenAI-compatible APIdocs
sglang serve datalab-to/surya-ocr-2.5 --host 0.0.0.0 --port 30000

Needs SGLang’s ROCm build (Instinct MI300-class GPUs).

Transformers · Ask about an imagedocs
from transformers import pipeline

pipe = pipeline("image-text-to-text", model="datalab-to/surya-ocr-2.5", 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"])

OCR models that fit the Radeon RX 9070 XT comfortably

All that fit →
ModelParamsBest quantMemoryLicence
stepfun-ai logoGOT-OCR2_0stepfun-ai
716MFP16 / BF16≈2.1 GBFits in memoryCommercial OK
PaddlePaddle logoPaddleOCR-VL-1.6PaddlePaddle
959MFP16 / BF16≈2.6 GBFits in memoryCommercial OK
PaddlePaddle logoPaddleOCR-VLPaddlePaddle
959MFP16 / BF16≈2.6 GBFits in memoryCommercial OK
tencent logoHunyuanOCRtencent
1.1BFP16 / BF16≈3.0 GBFits in memoryConditions
777MFP16 / BF16≈2.2 GBFits in memoryCommercial OK
datalab-to logosurya-ocr-2datalab-to
686MFP16 / BF16≈2.0 GBFits in memoryConditions

Other GPUs that run surya-ocr-2.5

Frequently asked questions

Can the Radeon RX 9070 XT run surya-ocr-2.5?
Yes. The Radeon RX 9070 XT runs surya-ocr-2.5 fully in VRAM at FP16 / BF16 (about 2.0 GB of 16 GB).
Which quantisation of surya-ocr-2.5 should I use on the Radeon RX 9070 XT?
FP16 / BF16: the largest one that fits. Expected quality: lossless compared with the full-precision weights.

Memory and speed are estimates from the model's size and the card's memory bandwidth; real numbers depend on the runtime, drivers and context length. Community figures are medians of user reports.