Features
Unified multimodal understanding and generation in one model
Autoregressive framework with unified Transformer architecture
Decoupled visual encoding pathways (SigLIP-L + MLP adapters)
Text-to-image generation with instruction following
Image understanding (description, analysis)
Available in 1B and 7B parameter variants
MIT license, open-source on Hugging Face and GitHub
WebGPU support for in-browser inference (1B model)
Compatible with ComfyUI via community nodes
Bidirectional image-text interaction
Optimized training with extended datasets and stability techniques
Model sizes: 1.3B (Janus), 1.3B (JanusFlow), 1B, 7B checkpoints
Sequence length 4096 for all models
Integration with Hugging Face and GitHub for deployment
Supports unrestricted commercial use
Chat with 600+ AI models (GPT-5.5, Claude Opus 4.8, Gemini 3.5 Flash, Grok 4.3)
Image generation (Nano Banana Pro, GPT-5 Image, FLUX.2 Pro, Seedream 4.5)
Video generation (Sora 2, Veo 3.1, Seedance 2.0, Kling 2.6, PixVerse 5.5)
Web search with citations (Sonar Pro)
Upload and chat with any file (no size limit)
Custom AI agents and assistants for workflows
Prompt library and builder for reusable prompts
Voice chat (dictation and text-to-speech)
Multi-model side-by-side comparison
Model fallback (automatic provider switching)
Usage dashboard for tracking consumption
Projects – organize chats with file sources and instructions
OpenAI-compatible API for external tools
MCP integrations (8000+ tools including Gmail, Slack)
Kimi K2.7 Code model for coding tasks (since June 2026)
