Chutes AI
Chutes is the leading open-source, decentralized compute provider for deploying, scaling and running open-source AI models in production, built on the Bittensor network. The platform democratizes AI infrastructure by making deployment as easy as deploying a website.
At a Glance
- AI and ML engineers building production applications
- Startups needing scalable AI infrastructure without high costs
- Researchers requiring powerful GPU compute
- Enterprise companies seeking serverless AI deployment
- +5 more
AI Tools by Chutes AI
(1)
Chutes AI
Serverless GPU Inference Platform
Discussions
No discussions yet
Be the first to start a discussion about Chutes AI
Latest News
chat.chutes.ai Platform Launch with 60+ Models and Mid-Conversation Model Switching
Fictio Character Roleplay Platform Launch with Creator Contests
$3 Per Month Starter Plan Announced for Accessible AI Compute
Chutes Wrapped 2025: 668K Users, 29.7 Trillion Tokens, $1.3M Revenue
Products & Services
Core serverless AI compute platform for deploying, scaling and running open-source models in production. Provides high-performance inference, ephemeral jobs, batch processing, and long-running tasks on decentralized GPU infrastructure.
Consumer chat application featuring over 60 AI models with unique mid-conversation model switching capability, allowing users to seamlessly change models during a conversation.
Consumer application for creative AI workloads including image, video, speech, music, and 3D generation tasks.
Confidential compute solution for AI inference with hardware-protected privacy, verifiable security, and 'don't trust, verify' security model. Enables running proprietary models on decentralized infrastructure with end-to-end encryption.
Market Position
Chutes positions itself as the leading open-source, decentralized AI compute provider with 85% lower costs than AWS and 70-85% cheaper than OpenAI. Example: DeepSeek R1 costs $0.30 per million input tokens on Chutes vs $10 on OpenAI (90% savings). Built on Bittensor's decentralized network (Subnet 64) with access to hundreds of H200s, B200s, and A6000 GPUs processing trillions of tokens monthly. Differentiators include: truly serverless deployment with no infrastructure management, confidential compute via TEE, 150+ open-source models, and auto-scaling from zero to hundreds of instances. Became a main provider on OpenRouter and achieved product-market fit with 668K users and 29.7 trillion tokens in first year.
Leadership
Founders
Namoray
Co-founder of Rayon Labs, a Bittensor-focused company that created Chutes. Leader in the Bittensor community managing multiple subnets including Chutes (SN64), Gradients (SN56), and Nineteen (SN19).
BonOliver
Co-founder of Rayon Labs and Frontend Developer at Chutes. Part of the decentralized collective that built and operates Chutes on the Bittensor network.
Jon Durbin
Renowned AI developer and creator of popular Hermes AI models and Airoboros instruction-tuning datasets. Expert in LLM fine-tuning and synthetic data generation. Received open-source AI grants for GPU-poor developers. Backend Developer and Founder of Subnet 64 (Chutes).
Executive Team
Jon Durbin
Founder / Backend Developer
Creator of Hermes AI models and Airoboros datasets, expert in instruction-tuning LLMs
Namoray
Founder (via Rayon Labs)
Co-founder of Rayon Labs, managing multiple Bittensor subnets
Founding Story
Chutes was founded in 2024 by a global, decentralized collective of engineers, researchers, and builders through Rayon Labs, with the vision of democratizing AI infrastructure. The founders believed AI should be accessible to every developer, researcher, and innovator - not just tech giants with unlimited resources. Built on the Bittensor network as Subnet 64, Chutes launched in late January 2025 to provide serverless, decentralized GPU compute for deploying and scaling open-source AI models. The platform emerged from the need to make AI deployment as simple as deploying a website, eliminating the infrastructure hassle and high costs of traditional cloud providers.
Business Model
Revenue Model
Subscription-based pricing tiers ($3-$20/month) plus pay-as-you-go (PAYG) usage beyond limits. Enterprise custom billing. Payments accepted in TAO tokens or fiat. Revenue is automatically staked to buy back subnet native tokens and reward miners and validators in the decentralized network. Cost efficiency: 85% cheaper than AWS, 70-85% cheaper than OpenAI.
Pricing Tiers
300 requests/day, unlimited API keys, unlimited models, access to Chutes Chat and Studio, PAYG requests beyond limit
2,000 requests/day, unlimited API keys, unlimited models, access to Chutes Chat and Studio, PAYG beyond limit, email support
5,000 requests/day, unlimited API keys, unlimited models, access to Chutes Chat and Studio, PAYG beyond limit, priority support
Unlimited API keys and models, access to Chutes Chat and Studio, dedicated support, SLA guarantees, private endpoints via VPC peering
Target Markets
- AI and ML engineers building production applications
- Startups needing scalable AI infrastructure without high costs
- Researchers requiring powerful GPU compute
- Enterprise companies seeking serverless AI deployment
- Developers building AI-powered products and applications
- Content creators using generative AI tools
- LLM chat applications and conversational AI
- Image and video generation services
- Real-time AI APIs at scale
- Batch processing workflows
- Model inference and deployment
- AI model training and fine-tuning
- OpenRouter
- Cline
- Roo Code
- Kilo / KiloCode.ai