🚀 GPT-6.2 Astra & Sol are now on Modellix.
• Astra: complex reasoning, coding & professional work
• Sol: everyday coding & agent workflows
Model IDs: openai/gpt-6.2-astra openai/gpt-6.2-sol
Try them through the Modellix LLM Gateway.
58
An octopus takes over the piano. Yeah, we made this with one still frame. 🐙
Meet Kling 4.0 on Modellix: image-to-video generation, now up to 4K resolution.
What would you bring to life?
🤖 Made with AI
1
105
More video for your budget. 🚀
Seedance 2.5 Flash is now on Modellix: near Seedance 2.5 quality, nearly 2× faster generation, and about 35% lower cost per second than Seedance 2.5.
Your next idea costs less to try. What will you make?
🤖 Made with AI
68
🚀 Qwen4 Max is now on Modellix.
• 1M-token context
• Text, image & video input → text output
• Thinking on by default
Pin this version: qwen/qwen4-max Follow the latest: ~qwen/qwen-latest
Try it through the Modellix LLM Gateway.
81
🚀 Create more, wait less.
Seedance 2.5 Flash is now live on Modellix!
What it supports:
• 🎬 Text-to-video with native audio
• 🖼️ Image-to-video with first/last frame control
• 🎨 Reference-to-video with up to 10 images
• 🎞️ Video-to-video editing and extension
• ⚡ 4–30s videos at 480p or 720p
Use the model IDs:
• bytedance/seedance-2.5-flash-t2v
• bytedance/seedance-2.5-flash-i2v
• bytedance/seedance-2.5-flash-r2v
• bytedance/seedance-2.5-flash-v2v
Start creating through the Modellix API.
🤖 Made with AI
🚀 Seedream 5.5 is now on Modellix.
Create with native transparent PNG output. Decompose an image into a base image and up to 99 transparent PNG layers.
🎨 Less flat. More flexible.
Try Seedream 5.5 on Modellix.
🎬 Gemini Omni 1.2 Flash is now on Modellix.
Generate videos with synchronized audio from text. Guide new clips with visual references. Edit, extend, or upscale existing footage.
One series. More ways to create.
Try it on Modellix.
2
105
🎬 Kling 4.0 & Kling 4.0 Flash are now on Modellix.
Guide image-to-video with first, middle & last frames in Kling 4.0—or start from a single frame with Flash.
Both support text-to-video with generated audio and reference-guided video editing.
Try them on Modellix.
3
129
Some deliveries should never be accepted. 🕷️
Seedance 2.5 generated this supernatural action short on Modellix.
-Stable spatial logic
-Natural physical collisions
-Integrated VFX
Try it on Modellix.
🤖 Made with AI
A desert world where every drop is worth the crossing. 🌌
Five cinematic sci-fi scenes, created with ChatGPT Image and brought to life with Vidu Q4 on Modellix
● Cinematic desert keyframe creation with ChatGPT Image
● Image-to-video generation with Vidu Q4
● 2K clips, up to 16 seconds per generation
● Prompt-guided motion across salt flats, a water vault, and an approaching dust storm
Edited into one film with an original score
Which scene would you step into: the salt flats, the water vault, or the storm?
What waits beyond the gothic cathedral gates? 🕯️
Haunting unbroken‑take gothic fantasy scenes built with Vidu Q4 Preview on Modellix.
Locked reference details
Smooth camera movement
Evocative mood lighting.
Try it on Modellix now.
🤖 Made with AI
🚀 Gemini 4 Hydrogen is now on Modellix.
Bring text, images, audio, video, and PDFs into everyday coding and agent workflows—with a 1M-token context window shared across input and output.
💻 Working with requirements, UI screenshots, or technical documents? Bring that context into your coding tasks, with text output for your next step.
📌 Model ID:
google/gemini-4-hydrogen
Try Gemini 4 Hydrogen through the Modellix LLM Gateway.
🪙 One coin to start. One coin to seal.
Urban fantasy showdown generated in Seedance 2.5 on Modellix.
-Consistent characte
-Fluid action
-Immersive sound
Try it on Modellix
🤖 Made with AI
2
221
🚀 Gemini 4 Carbon is now on Modellix.
Bring text, images, audio, video, and PDFs into complex technical workflows—with a 1M-token context window shared across input and output.
🧠 Built for complex reasoning and long-horizon agentic work, spanning software engineering, cybersecurity, and scientific reasoning.
📌 Model ID: google/gemini-4-carbon
Try Gemini 4 Carbon through the Modellix LLM Gateway.
🚀 Claude Fable 5.2 is now live on the Modellix LLM Gateway.
• 1M-token context window
• Text and image input, with text output
• Built for complex reasoning and long-horizon agentic work
Working through long documents, screenshots, or multi-step agent tasks? Try it with your own workflow.
📌 Keep your integration on 5.2 with the fixed model ID:
anthropic/claude-fable-5.2
🔄 Automatically follow the latest Fable model with the rolling alias:
~anthropic/claude-fable-latest
👉 Try Claude Fable 5.2 through the Modellix LLM Gateway.
1
154
📣 Six OpenAI Live models are now available on Modellix.
What’s included:
• 🎙️ Realtime voice — GPT Realtime 2.1, GPT Realtime 2.1 Mini, and GPT Live 1
• 📝 Realtime transcription — GPT Live Transcribe and GPT Realtime Whisper
• 🌐 Streaming speech-to-speech translation — GPT Realtime Translate
Use these model IDs:
• openai/gpt-realtime-2.1
• openai/gpt-realtime-2.1-mini
• openai/gpt-live-1
• openai/gpt-live-transcribe
• openai/gpt-realtime-whisper
• openai/gpt-realtime-translate
All six models are available through the Modellix Live WebSocket gateway.
🤖 Made with AI
Same prompt. Two models. Two very different sheep universes 🐑
We gave GPT Image 2 and Nano Banana 2.1 the exact same creative prompt—and the results surprised us.
GPT Image 2 pushes the depth and poster-like composition, while Nano Banana 2.1 leans into a softer, more playful character style.
Which approach gets your vote?
Created on Modellix.
🤖 Made with AI











