Sensenova/SenseNova-Loop-9B-Preview · Hugging Face
The model processes text and images, generating text for applications combining document understanding, coding, and external tool interactions. Its architecture includes a looped decoder with two-stream manifold-constrained hyper-connections (mHC), where a 24-layer block is executed twice with shared weights, effectively doubling the computation path to 56 layers while reusing only 32 distinct…
Key points
- Uses a looped decoder with 24-layer blocks executed twice, doubling computation to 56 layers
- Supports text, images, code, and tool trajectories with sequences up to 256K tokens
The model supports MTP for accelerated inference, with Dspark and Dflash2 under development. Deployment requires a documented environment and configuration. The model is licensed under Apache License 2.0 and released via Hugging Face. Downloads last month reached 12.
sensenova/SenseNova-Loop-9B-Preview · Hugging Face
huggingface.co · 30 October 2026
Loading the full article…
This text was published by huggingface.co. It is reproduced here with attribution so you can read it in full; the rights remain with the publisher. Read it at the source ↗
Coverage and discussion
1source- Reddit discussionreddit.com
The headline, key points and digest above were generated by Digest AI's editorial model from the linked sources. Automated summaries can contain errors: the sources are the record. Spotted a mistake? Tell us. Published by Martin K., who runs Digest AI and handles corrections.
More in Generative AI & Models
All →- Google’s Gemini 4 Carbon ranks #1 on Artificial Analysis Intelligence Index v5 with score of 59 · 4 src
- NVIDIA releases open-weight Nemotron 3.5 Transcribe model and Granary v2 dataset · 12 src
- OpenAI launches GPT-6.1 Terra for faster, better-performing AI agents at half the price · 19 src
- Baseten releases 2B-parameter decision model joey-1 at $0.01 per million tokens · 2 src
- Artificial Analysis ranks Claude Opus 5.5 first in updated Intelligence Index v5 · 2 src
Comments
via GitHub Discussions