Now happening in VibePaper
Product news, model capabilities, and creator stories.
Let Agent carry every step from intent to delivery
Models, tools, and workflows collaborate on one canvas.
MODEL INFRASTRUCTURE · PIXELENGINEModel capabilities by Nexra
Explore nowModels, tools, and workflows collaborate on one canvas.
MCP & CLIStart from Codex and Claude Code
Seedance 2.0High-quality video generation from multimodal references
Nano Banana 2 LiteFast generation and editing with multiple references
PaperTVDiscover and explore AI-created videos
GPT-Image-2High-quality image generation and precise editing
Start with a distinctive Skill and create with confidence
Explore PaperSkill
Discover remarkable work and shape what’s next
Explore PaperTV
Models
One workspace. Every leading model.
Routed automatically by task, with human preferences preserved. New models hot-plug in without touching your flow.
Text
Gemini 3.1 ProReasoning
Gemini 3.5 FlashFast & light
GPT-5.5Tool use
Claude Opus 4.8Deep analysis
Claude Sonnet 4.6Balanced
Claude Haiku 4.5Fast Claude
Grok 4.3Vision reasoning
Kimi K2.5Multimodal
Seed 2.0 ProChinese writing
Seed 2.0 MiniCompact
Image
Nano BananaFast image
Nano Banana 2High fidelity
Nano Banana ProQuality & speed
Z-Image TurboFast generation
Qwen Image Edit PlusQwen editing
Seedream 5.02K/4K
Grok ImaginexAI quality
Midjourney V8.1Signature style
GPT Image 2Flexible sizes
Wan 2.7 Image ProAspect control
Video
Veo 3.1Film look
Veo 3.1 LiteLite & fast
Kling 3.0 OmniCinematic
Seedance 2.0Live-action
Seedance 2.0 FastFast video
Wan 2.7Steady
HappyhorseAliyun video
Vidu q3 ProReference video
MiniMax Hailuo 2.3 FastFirst/last frame
PixVerse V6VFX specialist
Audio / World
ElevenLabs Multilingual V2Multilingual voice
Minimax Speech 2.8Emotive speech
Fish Audio S2 ProHigh quality audio
Doubao TTS V2Chinese TTS
Marble 1.1 PlusWorld generation
FAQ
Frequently asked
Do I need to write code?
No. Describe your goal in natural language and the workflow compiles into an editable node graph. Step in only when you want to refine.
Who owns the generated content?
You own everything you create on VibePaper, free to use commercially.
Which models are supported?
Text, image, video, voice and research connect to leading models. New models hot-plug in without changing your flow.
Can my team collaborate?
Workflows and outputs are shareable — teammates can work together on the same canvas in real time.





