Run PaddlePaddle/PaddleOCR-VL-1.6-GGUF locally

License: apache-2.0 ⬇ 611,348 ❤ 32
Parameters0.47B
Context131,072

PaddlePaddle/PaddleOCR-VL-1.6-GGUF is a compact language model with 0.47 billion parameters, built on the paddleocr architecture. It is released under the apache-2.0 license and has been downloaded 611,348 times.

To run PaddlePaddle/PaddleOCR-VL-1.6-GGUF locally at a 4,096-token context, its quantized versions need between 2.97 GB (GGUF, lowest quality) and 2.97 GB (GGUF, highest quality) of memory, weights plus KV cache and a system margin included.

For most users the best balance is GGUF, needing about 2.97 GB. That means PaddlePaddle/PaddleOCR-VL-1.6-GGUF fits entirely in the VRAM of a 6 GB GPU or larger, running fully on the GPU.

→ Guide: How much VRAM do you need?

All quantizations

Quant.Bits QualityWeights KVTotal Speed~Verdict
GGUF 31.16 Excellent 1.69 GB 0.47 GB 2.97 GB 236.3 t/s Fits in VRAM

KV cache computed from the model's exact architecture. Speed is a rough estimate bounded by memory bandwidth.

Frequently asked questions

How much VRAM do you need to run PaddlePaddle/PaddleOCR-VL-1.6-GGUF?

You need about 2.97 GB of VRAM to run PaddlePaddle/PaddleOCR-VL-1.6-GGUF entirely on the GPU using the GGUF quantization (at a 4,096-token context). Smaller quantizations lower the requirement at the cost of quality.

Can I run PaddlePaddle/PaddleOCR-VL-1.6-GGUF on an 8 GB GPU?

Yes. With 8 GB of VRAM you can run PaddlePaddle/PaddleOCR-VL-1.6-GGUF fully on the GPU using GGUF (about 2.97 GB).

Can I run PaddlePaddle/PaddleOCR-VL-1.6-GGUF on a 16 GB GPU?

Yes. With 16 GB of VRAM you can run PaddlePaddle/PaddleOCR-VL-1.6-GGUF fully on the GPU using GGUF (about 2.97 GB).

Can I run PaddlePaddle/PaddleOCR-VL-1.6-GGUF on a 24 GB GPU?

Yes. With 24 GB of VRAM you can run PaddlePaddle/PaddleOCR-VL-1.6-GGUF fully on the GPU using GGUF (about 2.97 GB).

What is the best quantization for PaddlePaddle/PaddleOCR-VL-1.6-GGUF?

If memory allows, higher bits-per-weight means better quality. A common sweet spot is a Q4_K_M or Q5_K_M quantization, which keeps most of the quality while roughly halving the memory versus 8-bit. Pick the highest quantization that still fits in your VRAM.