Interfaze 1 Lite
Interfaze 1 Lite is an open-weight mixture-of-architectures model for developer workloads, combining a vision-language reasoning core with specialist components. It supports document understanding, speech transcription and diarization, visual grounding, structured output, translation, forecasting, and safety checks.
What it does
A person can use it to extract structured information from documents, transcribe and diarize audio, detect or ground objects and interface elements, forecast time series, and moderate text.
- Availability
- Download from Hugging Face or use the Interfaze API.
- Pricing model
- Not established
- Deployment
- Runs offline on one 80 GB GPU with no external services.
- Privacy
- Runs offline with no external services.
- Licence
- Apache-2.0
Capability
Modality
Access
Deployment
Pricing
Target user
Evidence
6 claims- The model reads text, reading order, tables, and layout from images, PDFs up to 50 pages per call, and Word files. [1]
- **Document understanding.** Text, reading order, tables and layout from images, PDFs (up to 50 pages a call) and Word files, with a box and confidence for every line and word.
Hugging Face. Quoted text is outside CC BY 4.0 and keeps its original rights. - Speech transcription supports timestamps and speaker diarization. [1]
- **Speech.** Transcription with timestamps and speaker diarization.
Hugging Face. Quoted text is outside CC BY 4.0 and keeps its original rights. - The model supports open-vocabulary object detection and GUI element grounding. [1]
- **Visual grounding.** Open-vocabulary object detection with outlines, and GUI element grounding for computer-use agents.
Hugging Face. Quoted text is outside CC BY 4.0 and keeps its original rights. - Responses can be constrained to a supplied JSON schema. [1]
- **Structured output.** Responses constrained to a JSON schema you supply, reading from any mix of text, images, documents and audio.
Hugging Face. Quoted text is outside CC BY 4.0 and keeps its original rights. - The whole model runs on one 80 GB GPU with no external services. [1]
The whole model runs on one 80 GB GPU with no external services.
Hugging Face. Quoted text is outside CC BY 4.0 and keeps its original rights. - The model can be downloaded from Hugging Face and run with Transformers, or accessed through the Interfaze API. [1]
Sources
1
1
Hugging FaceOwner-controlled
https://huggingface.co/interfaze-ai/interfaze-1-lite
Retrieved 2026-10-28 00:51 UTC
Verified
Versions
5| Version | Effective | Change | Changed fields | State |
|---|---|---|---|---|
| 5 | 2026-10-14 12:46 UTC | material change | capability, modality, access, availability, deployment | Current |
| 4 | 2026-10-12 14:35 UTC | material change | capability, modality, access, deployment, availability, deployment | Current |
| 3 | 2026-10-08 18:11 UTC | material change | capability, modality, access, deployment, availability, deployment | Current |
| 2 | 2026-10-06 20:09 UTC | material change | capability, modality, deployment, availability, deployment, privacy | Current |
| 1 | 2026-10-05 21:08 UTC | publication | None | Current |
In briefs
1- Build and govern workplace agents with North 2 Daily brief 2026-10-05
Original catalog content is licensed under CC BY 4.0, attributed to wowsuchai. Source quotations and linked content keep their original rights.