FA
Fara1.5-27B
microsoft
ID: fara15-27b
open weights·Active·multimodal general·Updated Jul 17, 2026
Fara1.5-27B is a 27.4B-parameter image text to text model developed by microsoft. Built on the Qwen3_5ForConditionalGeneration architecture using transformers. Supports en language(s). Released on 2026-07-17 with 122 likes and 1,406 downloads on Hugging Face.
Fara1.5-27B
Model Overview
Fara1.5-27B is a 27.4B-parameter image text to text model developed by microsoft. Built on the Qwen3_5ForConditionalGeneration architecture. Supports en. Released on 2026-07-17.
📊 Quick Specs
Specification Table
| Specification | Value |
|---|---|
| Parameters | 27.4B |
| Architecture | Qwen3_5ForConditionalGeneration |
| Task | image text to text |
| Modality | text, image |
| License | MIT |
| Framework | transformers |
| MoE | No |
| Languages | en |
✨ Key Features
- 27.4B parameters
- Built on Qwen3_5ForConditionalGeneration architecture (transformers)
- Primary task: image text to text (text, image modality)
- Language support: en
- Open-weights under MIT license — self-hostable and fine-tunable
📈 Community Adoption
- 122 likes on Hugging Face
- 1,406 downloads on Hugging Face
🔗 Resources
- Hugging Face Hub: Fara1.5-27B on Hugging Face
- Paper: arXiv
📜 License & Access
MIT — Open-weights model available for download, fine-tuning, and self-hosted deployment.