Model provider · Bring your own keyCerebras in a visual AI workflow
Run open models on Cerebras' wafer-scale inference in NodeTool — extreme tokens-per-second chat, called with your own Cerebras key.
Cerebras serves open models on wafer-scale hardware for some of the fastest inference available. In NodeTool it's a low-latency backend for chat and agent nodes that run the same open models.
Cerebras is bring-your-own-key: set `CEREBRAS_API_KEY` and NodeTool calls Cerebras directly at list price.
Calls go straight to Cerebras with your key at their list price — NodeTool never proxies or marks up the request.
What Cerebras does in NodeTool
Featured Cerebras models
NodeTool queries Cerebras for the exact models your key can access and exposes each as a node. These are the families you can build with.
Open Llama models at extreme tokens/sec.
Popular open models on wafer-scale hardware.
Frequently asked
- Why use Cerebras in NodeTool?
- Latency. Cerebras serves open models extremely fast, which speeds up agent loops and interactive chat nodes.
- How do I connect Cerebras?
- Add `CEREBRAS_API_KEY` in settings and pick a Cerebras model on a chat or agent node.
- What does Cerebras cost in NodeTool?
- Cerebras' list price — NodeTool adds no fee.
Other providers
fal.ai
Replicate
Kie
Together AI
AtlasCloud
Topaz
Anthropic
OpenAI
Google Gemini
xAI Grok
Mistral AI
Groq
DeepSeek
Cohere
OpenRouter
Hugging Face
Moonshot (Kimi)
ElevenLabs
Reve
Meshy
Rodin (Hyper3D)
MiniMaxRun Cerebras your way.
Download NodeTool Studio and build across image, video, audio, and text with your own keys.