Home/Events/Hugging Face Releases Multimodal Open d1 Decision Models: d1-3B and d1-omni-600M

Hugging Face Releases Multimodal Open d1 Decision Models: d1-3B and d1-omni-600M

Confirmed

Confidence

90%

Impact: 80%

Updated Oct 7

Consensus Brief

Hugging Face has introduced two new open decision models, d1-3B and d1-omni-600M, which support multimodal inputs including text, images, and audio. The d1-3B model is noted for its high performance, scoring 48.57 on the Decision Index 0.2.1, making it the best decision model under 10B parameters.

Sourced from

Hugging FaceT1

Primary: Hugging Face

What Changed Since Last Update

Oct 7

The release of d1-3B and d1-omni-600M marks the introduction of new models in the d1 decision model family, enhancing capabilities for edge inference with multimodal support.

Claim Ledger

4 claims tracked across sources

Confirmed Fact

d1-3B scores 48.57 on the Decision Index 0.2.1, outperforming all 4B and 9B models.

Confirmed Fact

d1-3B answers a question in 16 ms on an NVIDIA Jetson AGX Thor.

Confirmed Fact

d1-omni-600M supports text and images or text and audio.

Confirmed Fact

d1-3B achieves a mean score of 82.9 across seven public datasets.

Role-Based Impact Analysis

Source Timeline

1 source corroborating

Multimodal open d1 decision models for the edge

Hugging Face·Oct 7