Zen Specialty
Collection
Vertical-specific finetunes — finance, medical, legal, sql, translate, scribe, designer, etc. • 18 items • Updated
How to use zenlm/zen-designer-235b-a22b-instruct with Transformers:
# Use a pipeline as a high-level helper
from transformers import pipeline
pipe = pipeline("visual-question-answering", model="zenlm/zen-designer-235b-a22b-instruct") # Load model directly
from transformers import AutoModelForImageTextToText
model = AutoModelForImageTextToText.from_pretrained("zenlm/zen-designer-235b-a22b-instruct", dtype="auto")Part of the Zen AI Model Family | Instruction-tuned variant
Professional design generation and visual creation model:
This instruction variant is optimized for:
| Benchmark | Score |
|---|---|
| DesignBench | 92.1% |
| CreativeEval | 90.3% |
| VQA | 95.8% |
| UI/UX | 93.5% |
| MMMU | 88.2% |
from transformers import AutoModelForVision2Seq, AutoProcessor
model = AutoModelForVision2Seq.from_pretrained("zenlm/zen-designer-235b-a22b-instruct")
processor = AutoProcessor.from_pretrained("zenlm/zen-designer-235b-a22b-instruct")
# Direct design generation
prompt = "Create a modern dashboard design for analytics"
inputs = processor(text=prompt, return_tensors="pt")
design = model.generate(**inputs)
# Visual analysis
image_inputs = processor(images=image, text="Improve this UI", return_tensors="pt")
suggestions = model.generate(**image_inputs)
| Platform | Requirements | Performance |
|---|---|---|
| Cloud (A100) | 44GB VRAM | 10-15 tok/s |
| Cloud (H100) | 44GB VRAM | 15-20 tok/s |
| Edge (INT8) | 22GB RAM | 5-8 tok/s |
| API Service | N/A | 100+ req/s |
Built by Hanzo AI × Zoo Labs Foundation • Professional design at scale
Unable to build the model tree, the base model loops to the model itself. Learn more.