Downloads · 30 days
275
31% of all-time downloads
surya-ocr-2-mlx
aglaia-models/surya-ocr-2-mlx
surya-ocr-2-mlx is a machine learning model from aglaia-models. Use it for the machine learning task on the model card, and read the license before you ship it in a product. It is set up for mlx. The card lists the license as openrail.
MLX conversion of datalab-to/surya-ocr-2 (a Qwen3.5/Qwen3-Next OCR VLM), runnable on Apple Silicon via mlx-vlm.
All-time downloads
881
Public
Parameters
666M
1.1 GB on disk
Likes
3
Public
Repo makeup
Click a slice to open those files.
.safetensors1 GB · 97%
Parameter types
How the weights are stored.
BF16357M · 54%
Base models
Model card
From the Hugging Face model README
Surya OCR 2 — MLX (mixed)
MLX conversion of datalab-to/surya-ocr-2
(a Qwen3.5/Qwen3-Next OCR VLM), runnable on Apple Silicon via
mlx-vlm.
Variant mixed. See the project's bench/RESULTS.md for the precision
benchmark. The mixed variant keeps the vision tower, embeddings, lm_head and the
18 Gated-DeltaNet layers at bf16 and quantizes only the bulk linears — strictly
more precise than the existing 8-bit port.
from mlx_vlm import load, generate
model, processor = load("aglaia-models/surya-ocr-2-mlx")
Changes from the original
Weights converted to MLX format and quantized (mixed). No fine-tuning.
License & attribution
Derivative of datalab-to/surya-ocr-2, under the AI Pubs OpenRAIL-M License
(Modified). This derivative is released under the same license (share-alike); the
use-based restrictions carry over. Credit: Datalab (https://github.com/datalab-to/surya).
- Library
- mlx
- Type
- qwen3_5
- License
- openrail
- Created
- Jun 30, 2026
- Updated
- Jun 30, 2026