HiDream I1 Models by HiDream AI
1 model2025
Last refreshed 2026-06-29. Next refresh: weekly.
Details
ResearcherHiDream AI
LicenseMITOSI-approved
Commercial useCommercial use: permitted
Models1
Released2025
Capabilities
MultimodalAll models
About
HiDream AI's open-source image generation model family using a sparse Diffusion Transformer with dynamic Mixture-of-Experts architecture. HiDream-I1 achieves state-of-the-art generation quality with 17B parameters at MIT license, available in Full, Dev, and Fast variants.
Decision facts
- Best fit
- image generationvision and multimodal work
- Capability starting point
- HiDream I1 with multimodal inputs
- Lowest tracked input
- Not tracked
- Closest related family
- 360Zhinao
Current Variants
Use-when guidance is based on each model's tracked capabilities, context window, release date, and replacement status.
1 in view
HiDream I1Current
Use when the workload needs image generation, 17B parameters, and multimodal inputs.
2025-04image generation17B parametersmultimodal inputs
| Model | Use when | Released | Signals | Status |
|---|---|---|---|---|
| HiDream I1 | Use when the workload needs image generation, 17B parameters, and multimodal inputs. | 2025-04 | image generation17B parametersmultimodal inputs | Current |
Release Timeline
1 release group2025-04
1 current
HiDream I1
Currentimage generation17B parametersmultimodal inputs
Specifications(1 models)
| Model | Released | Parameters | Multimodal |
|---|---|---|---|
| HiDream I1 | 2025-04 | 17B | Yes |