
Build with your API key
Build a website with your NVIDIA NIM API key
Facts last verified: . Provider pricing and free tiers change — check their site for current numbers.
Yes — NVIDIA's NIM API serves GPU-optimized open models (DeepSeek R1, Nemotron) through build.nvidia.com, and new developer accounts include free credits. Create a key, paste it into OpenThorn (free, encrypted server-side), describe your site, and build your first pages at no cost.
Get your key
Official pricing
Free tier
Yes — developer accounts receive free credits on build.nvidia.com to evaluate models.
Recommended models
- DeepSeek R1 — Strong reasoning model served on NVIDIA infrastructure.
- Llama 3.1 Nemotron Ultra 253B — NVIDIA's tuned large model — capable for complex builds.
How to build your website
- Create an NVIDIA API keySign up at build.nvidia.com — developer accounts include free credits — and generate an API key from the model catalog.
- Pick a model from the catalogBrowse build.nvidia.com for available models; DeepSeek R1 and Nemotron are strong choices for code.
- Connect the key to OpenThornSign up free at openthorn.app, open the Providers page, choose NVIDIA NIM, and paste your key. It is encrypted server-side with AES-256-GCM and never exposed raw.
- Describe your website and ship itCreate a project, pick a model, and describe the site. Preview it live, iterate, then deploy to Netlify or export the source.
What it costs
Free developer credits cover first sites at $0; afterwards, open-model rates keep a typical website in the cents range.
Common questions
Can I use NVIDIA NIM with OpenThorn for free?
Yes — new developer accounts on build.nvidia.com include free credits, and OpenThorn is free, so your first website builds cost $0.
Which NVIDIA-served model is best for website generation?
DeepSeek R1 is the strongest for reasoning-heavy builds; Llama 3.1 Nemotron Ultra is NVIDIA's tuned large model and handles complete sites well.
Is my NVIDIA API key safe in OpenThorn?
Keys are encrypted server-side with AES-256-GCM using per-user derived keys and are never exposed raw to the browser. You can rotate the key on build.nvidia.com at any time.
New here? Read how to build a website with AI using your own key, how to get an AI API key, or compare model pricing. Other providers: OpenAI · Anthropic · Google Gemini · DeepSeek · Mistral AI · Groq · Together AI · xAI · Cohere · Perplexity · OpenRouter · Ollama · Fireworks AI · Cerebras · Azure OpenAI · Amazon Bedrock
