datalab-to models

3 datalab-to models from datalab-to on Hugging Face, from 686M to 9.7B parameters, published by datalab-to in BF16, with a 256K-token context. The smallest official model, surya-ocr-2, needs about 1.5 GB of VRAM at its published precision; the cheapest live fit is RTX 3070 at $0.088/hr.

Pick a size

One row per official datalab-to size: the VRAM it needs at each precision, the cheapest GPU that holds it today, and the KV cache for a 32K-token context.

ModelParametersNative VRAMFP8 VRAMINT4 VRAMLive GPU fit (native)Est. $/hrKV cache at 32K
surya-ocr-2686M1.5 GB0.8 GB0.4 GBRTX 3070$0.088/hr0.38 GB
chandra-ocr-25.3B11.8 GB5.9 GB3.0 GBRTX 5060 Ti$0.145/hr1.0 GB
lift9.7B21.6 GB10.8 GB5.4 GBRTX A5000$0.176/hr1.0 GB

VRAM is weights times a flat 1.2 overhead; the KV cache is a separate per-model figure at 16-bit, shown where the architecture is published. See the methodology.

Official models (3)

ModelParametersPublished asVRAM neededLive GPU fitEst. $/hr
surya-ocr-2686MBF161.5 GBRTX 3070$0.088/hr
chandra-ocr-25.3BBF1611.8 GBRTX 5060 Ti$0.145/hr
lift9.7BBF1621.6 GBRTX A5000$0.176/hr

VRAM is for the precision the model is published in: the weights times a flat 1.2 overhead, with the KV cache not included. See the methodology. The fit shown is the lowest-priced single GPU type that holds the model at that precision, or the lowest-priced multi-GPU set (up to 8) when none does.

Start ComfyUI on a rented GPU.

© 2026 Aquanode. All rights reserved.

All trademarks, logos and brand names are the property of their respective owners.