SmolVLM2-500M-Video-Instruct-ternary – AI Model by AsadIsmail | AlphaNeural AI

You can deploy this model and start earning money today!

AsadIsmail/
SmolVLM2-500M-Video-Instruct-ternary

like

0

image-text-to-text

HuggingFaceTB/SmolVLM2-500M-Video-Instruct

endpoints_compatible

Views

No views yet

SmolVLM2-500M-Video-Instruct — Ternary Quantized (tritplane3)

Ternary-quantized version of HuggingFaceTB/SmolVLM2-500M-Video-Instruct — a compact video-understanding VLM.

Specifications

Property	Value
Base Model	HuggingFaceTB/SmolVLM2-500M-Video-Instruct
Parameters	500M
Quantization	tritplane3 (225 linear layers)
Full-model effective bits	11.50
Compression ratio	1.39×
Avg reconstruction error	0.1413
Vision Encoder	FP16 (preserved)

Collection

Part of ternary-models.

GitHub: github.com/Asad-Ismail/ternary-models