Model comparison
Stable Diffusion (CompVis) vs GPT-5
Capability radar, value bars, and side-by-side posture.
Stable Diffusion (CompVis)
Not scored
value score
GPT-5
94
value score
Stable Diffusion (CompVis) ctx
—
GPT-5 ctx
400K
Capability radar
Value · context · multimodal · openness · speed posture
Value head-to-head
- GPT-594
Attribute tape
| Field | Stable Diffusion (CompVis) | GPT-5 |
|---|---|---|
| Developer | CompVis | OpenAI |
| Context | — | 400K |
| Modalities | Text, Code, Image, Audio | |
| Openness | — | Proprietary Model |
| Speed | — | Fast |
| Price | — | Mid-High |
| Value score | — | 94 |
| Summary | The original open-source text-to-image latent diffusion model released by LMU Munich's CompVis lab in collaboration with Stability AI and Runway. It generates photorealistic images from text prompts and runs on consumer GPUs, with weights and code publicly available. | OpenAI frontier multimodal model. |