I've been running some tests lately and I'm honestly blown away.
I just tried the new HunyuanOCR (specifically the GGUF versions) and the performance on budget hardware is insane. Using the 1B parameter model, I’m getting around 90 t/s on my old GTX 1060.
The accuracy is nearly perfect, which is wild considering how lightweight it feels.
I see a lot of posts here asking for reliable, local OCR tools that don't require a 4090 to run smoothly—I think this might be the missing link we were waiting for.
GGUF:
https://huggingface.co/ggml-org/HunyuanOCR-GGUF/tree/main
ORIGINAL MODEL:
https://huggingface.co/tencent/HunyuanOCR
[link] [comments]