axteam/GLM-OCR-FP8-dynamic
Image-Text-to-Text • 1B • Updated • 19 • 1
OCR Quantized
Quantized OCR / document VLMs (GLM-OCR, PaddleOCR-VL) — compressed-tensors + GGUF for vLLM / llama.cpp. By axteam.
Note FP8 dynamic (W8A8) — near-lossless 8-bit, vLLM-servable
axteam/GLM-OCR-GGUF
0.9B • Updated • 90Note GGUF Q8_0 / Q4_K_M (text backbone) for llama.cpp
axteam/PaddleOCR-VL-FP8-dynamic
Image-Text-to-Text • 1.0B • Updated • 16Note FP8 dynamic (W8A8) — near-lossless 8-bit, vLLM-servable
axteam/PaddleOCR-VL-GGUF
0.5B • Updated • 92 • 1Note GGUF Q8_0 / Q4_K_M (text backbone) for llama.cpp