Vercel AI Gateway
Fastest way to switch everything onEvery text model in the catalogue, OpenAI and Nano Banana images, ElevenLabs narration and HeyGen video, with one key and one bill.
AI_GATEWAY_API_KEY
Nio Forge routes each request to the provider that owns the model. Keys live in server environment variables and never reach the browser.
Agent and text generation
Checking
Image generation
Checking
Narration
Checking
Video generation
Checking
Live voice roleplay
Checking
Every text model in the catalogue, OpenAI and Nano Banana images, ElevenLabs narration and HeyGen video, with one key and one bill.
AI_GATEWAY_API_KEY
Claude models direct, with prompt caching and safety fallbacks.
ANTHROPIC_API_KEY
GPT models, image generation and live speech to speech voice roleplays.
OPENAI_API_KEY
Gemini models and Nano Banana image generation.
GEMINI_API_KEY
Studio quality narration billed to your own ElevenLabs account. Not needed when AI Gateway is connected.
ELEVENLABS_API_KEY
Fast, low cost image models such as FLUX.
FAL_KEY
An alternative single key route to every text model.
OPENROUTER_API_KEY
Accounts, cloud projects, media storage and the server side credit ledger.
NEXT_PUBLIC_SUPABASE_URL, NEXT_PUBLIC_SUPABASE_PUBLISHABLE_KEY, SUPABASE_SERVICE_ROLE_KEY
Paid plans and credit packs, with Paddle as merchant of record (tax and VAT handled).
PADDLE_API_KEY, PADDLE_CLIENT_TOKEN, PADDLE_WEBHOOK_SECRET, PADDLE_PRICE_*
One click publish from Forge into the Nio Learn library.
NIO_LEARN_API_URL, NIO_LEARN_API_KEY
Google Docs, Sheets and Slides as source material, and Google Slides imported as slides.
NEXT_PUBLIC_GOOGLE_CLIENT_ID, NEXT_PUBLIC_GOOGLE_API_KEY, NEXT_PUBLIC_GOOGLE_APP_ID
| Model | Provider | Use | Direct id | Gateway id |
|---|---|---|---|---|
| Claude Haiku 5.5 | anthropic | Text, fast | claude-haiku-5-5 | anthropic/claude-haiku-5.5 |
| GPT-6 Luna | openai | Text, fast | gpt-6-luna | openai/gpt-6-luna |
| Gemini 3.8 Flash | Text, fast | gemini-3.8-flash | google/gemini-3.8-flash | |
| Claude Sonnet 5.5 | anthropic | Text, balanced | claude-sonnet-5-5 | anthropic/claude-sonnet-5.5 |
| GPT-6.1 Sol | openai | Text, balanced | gpt-6.1-sol | openai/gpt-6.1-sol |
| Gemini 4 Hydrogen | Text, balanced | gemini-4-hydrogen | google/gemini-4-hydrogen | |
| Claude Opus 5.5 | anthropic | Text, max | claude-opus-5-5 | anthropic/claude-opus-5.5 |
| Claude Fable 5.2 | anthropic | Text, max | claude-fable-5-2 | anthropic/claude-fable-5.2 |
| GPT-6 Astra | openai | Text, max | gpt-6-astra | openai/gpt-6-astra |
| Gemini 4 Carbon | Text, max | gemini-4-carbon | google/gemini-4-carbon | |
| GPT Image 2.5 Flare | openai | image | gpt-image-2.5-flare | |
| GPT Image 2.5 Sunburst | openai | image | gpt-image-2.5-sunburst | |
| Nano Banana 2.1 | image | gemini-nano-banana-2.1 | ||
| FLUX Schnell (fal) | fal | image | fal-ai/flux/schnell | |
| Nano Banana Pro (fal) | fal | image | fal-ai/nano-banana-pro | |
| ElevenLabs Flash 2.5 | elevenlabs | speech | eleven_flash_v2_5 | elevenlabs/eleven-flash-v2.5 |
| ElevenLabs v4 | elevenlabs | speech | eleven_v4 | elevenlabs/eleven-v4 |
| OpenAI TTS (retires 2027-01-06) | openai | speech | gpt-4o-mini-tts | |
| GPT Realtime 2.1 mini | openai | Live voice | gpt-realtime-2.1-mini | |
| GPT Realtime 2.1 | openai | Live voice | gpt-realtime-2.1 | |
| HeyGen Video | heygen | video | heygen-video-1 | heygen/heygen-video-1 |
The catalogue lives in src/lib/ai/models.ts. Add a row there to offer a new model; prices drive the credit maths.