Every provider, bring your own key
NodeTool integrates 23 model providers — media aggregators serving 1,777+ image, video, audio, and 3D models, plus direct LLM and voice APIs. Every one is a node on the canvas, called with your own key at the provider's list price — no credits, no markup.
Model aggregators
Run fal.ai's image, video, audio, and 3D models in a visual AI workflow — the largest catalog in NodeTool, called with your own key at fal's list price.
Run Replicate's image, video, and audio models in a visual AI workflow — hundreds of community and first-party models, BYOK at Replicate's list price.
Run Kie's image, video, and audio models in a visual AI workflow — Veo, Kling, Seedream, Suno and more, called with your own key at Kie's price.
Run Together AI's image and video models in a visual AI workflow — FLUX, Seedream, Kling, Veo and more, BYOK at Together's list price.
Run AtlasCloud's image and video models in a visual AI workflow — Seedance, GPT-Image and more, called with your own key at AtlasCloud's price.
Run Topaz's image and video enhancement models in a visual AI workflow — upscaling and restoration, called with your own key at Topaz's price.
LLM & AI APIs
Run Anthropic's Claude models in NodeTool chat and agent nodes — long-context reasoning, vision, and tool use, called with your own Anthropic key.
Run OpenAI's GPT, reasoning, image, and audio models in NodeTool — chat, vision, GPT-Image, transcription, and TTS, called with your own OpenAI key.
Run Google's Gemini models in NodeTool — long-context multimodal chat with image and audio understanding, called with your own Gemini API key.
Run xAI's Grok models in NodeTool chat and agent nodes — capable reasoning with tool use, called with your own xAI key.
Run Mistral's chat, vision, and code models in NodeTool — efficient open-weight and commercial models, called with your own Mistral key.
Run open models on Groq's ultra-fast inference in NodeTool — very high tokens-per-second chat, called with your own Groq key.
Run DeepSeek's chat and reasoning models in NodeTool — strong reasoning at low cost, called with your own DeepSeek key.
Run Cohere's Command, Embed, and Rerank models in NodeTool — chat plus best-in-class retrieval for RAG, called with your own Cohere key.
Run hundreds of models through OpenRouter in NodeTool — one key routed to many providers, called at OpenRouter's price.
Run Hugging Face-hosted models in NodeTool — open models across text, image, and audio through Inference, called with your own HF token.
Run open models on Cerebras' wafer-scale inference in NodeTool — extreme tokens-per-second chat, called with your own Cerebras key.
Run Moonshot's Kimi models in NodeTool — long-context chat and agentic coding, called with your own Kimi key.
Media & 3D
Run ElevenLabs' voice models in NodeTool — text-to-speech, voice cloning, dubbing, and sound effects, called with your own ElevenLabs key.
Run Reve's image models in NodeTool — high-quality generation with strong prompt adherence and text rendering, called with your own Reve key.
Run Meshy's 3D models in NodeTool — text-to-3D and image-to-3D with textured meshes, called with your own Meshy key.
Run Hyper3D's Rodin models in NodeTool — image-to-3D generation of production-ready meshes, called with your own Rodin key.
Run MiniMax's video, chat, and speech models in NodeTool — including Hailuo video, called with your own MiniMax key.