All the models in one place
Run P-Image Upscale and every other model on Glif from one chat. One account, one balance, no API keys to wire up.
P-Image Upscale
How it did on the standard Glif tests
Every Glif test P-Image Upscale has run, straight from our eval suite. Open a test to see how every other model handled the same prompt, or browse the whole suite.
P-Image Upscale scorecard
Every output above was scored by gpt-5.5 against the capabilities each test is designed to probe.
Glif test scores
- Overall
- 4.8/10
- Prompt adherence
- 5.7/10
- Tests judged
- 3
- Overall rank
- #32/34