Service Interruption Notice
We experienced a service outage due to a storage issue with our server provider. Although the service is now partially restored, delays are still occurring. We are working diligently to fully resolve the issue as soon as possible.
We sincerely apologize for the inconvenience and appreciate your patience and understanding.
New Updates
- OpenAI's o4-mini and o3 are now available in AI Chatbot, while o3-mini has been deprecated.
- GPT-4.1, GPT-4.1 nano, and GPT-4.1 mini are now available in AI Chatbot.
- AI Chatbot and Story image settings are now replaced by 'Upload Files,' which now includes a preview and supports drag-and-drop with multiple image references.
- Latesync has been added to AI Video for lip-syncing videos with audio.
- Wan 2.1 I2V now runs on dedicated GPUs in AI Video, allowing for higher usage limits.
- New buttons under image messages: Convert to video, Re-generate image with GPT-4o Image
- New button under video messages: Lipsync video
- Switched from SSE to WebSocket for improved server and web communication stability.
- We sincerely apologize for any service interruptions experienced on April 16 due to this transition.
Shan Vem Apologies for the inconvenience. It worked well in our general testing environment, but we've identified an issue in your case from the logs. The content was generated successfully but didn’t appear in the chat. I'm looking into the issue now and have emailed you regarding this.
I appreciate on your responses.
New Updates
- FLUX.1 Dev has been added to AI Image with support for various tasks:
- Text to Image: Create images from text prompts.
- Image to Image: Transform existing images into new creative versions.
- Text to Image with LoRA: Generate images from text using fine-tuned LoRA models.
- Image to Image with LoRA: Modify images using LoRA-powered enhancements.
- Fill Inpaint: Repair, replace, or edit specific regions of an white masked image.
- Fill Outpaint: Expand image borders with AI-generated content that blends seamlessly.
- Redux Variation: Generate fresh variations of an existing image.
- ControlNet LoRA: Guide image generation with structural control using advanced LoRA models.
- See rate limits by plan here.
- A global CDN with 119 access points has been added, making loading speeds much faster.
- Helpful tooltips are now available for each AI mode setting.
- Mindmap downloads have been improved to an interactive HTML format for a better experience.
New Updates
- Llama 4 Maverick and Scout have been added to AI Chatbot.
- DiffRhythm is now available in AI Audio for music generation.
- MMAudio has been upgraded to V2 in AI Audio for video to audio.
- BEN has been upgraded to BEN2 in AI Image for background removal.
New Updates
- GPT-4o image generation is now in AI Chatbot as an experimental preview and may be unstable.
- The following models have been added to AI Video, available for all plans:
- Wan 2.1
- Luma Dream Machine
- AI Hug Video
- Kling has been added to AI Video. Available for Supporter plan and above.
The same problem has been observed for several days now. The AI bot can ask a maximum of 1-2 questions per hour, and it doesn't respond to further requests for an hour. It writes something like "GPT 4.1 is working". The bot can stay in this state for a very long time. I've tried different models for the same query, but nothing changes when changing models, the problem remains.
Yes, it’s the same delay issue. Our server provider HostHatch has had its storage server running slowly for the past few hours (not fully down but sluggish), so responses are delayed. We are migrating all data to a more reliable server as quickly as possible. I apologize for the inconvenience and thank you for your patience.
I have tested the video option. Its showing the circle not moving into making a video. Please check and confirm.