Deploy OSS models in minutes not days

    Start free

    No fee up to 730 GPU hours/month

    Kimchi platform model deployment setup interface

    Trusted by +2100 companies globally

    AkamaiBMW GROUPFICOHugging FaceShareChat
    G2 Awards - High Performer, Momentum Leader, Regional Leader Fall 2025

    Run models in confidence

    Deploy models

    • No manual provisioning
    • No infra overhead
    • Keep data private
    • Autoscale with ease

    Scale

    • Automatically to handle traffic spikes
    • No rate limits, full service uptime
    • Built-in cost optimisation

    Cost optimization

    • Share GPU nodes between models with MIG and time sharing
    • Seamlessly run on Spot GPU, up to 70% cheaper
    • No idle time, scale down to 0 intelligently

    See how Qwen2.5:14B can be cheaper than GPT-4o-Mini. Read the article →

    Pricing

    Free

    Start experimenting with Kimchi and test its value

    $0/month
    Start free

    No credit card needed

    Access to all Kimchi features
    Up to 10k gateway requests
    Up to 730 GPU hours/month

    Growth

    Run the gateway and model deployments in production at scale

    $500/month
    Get started
    Everything in Free
    Gateway - $5/100k events
    VPC deployments - $0.05/GPU hour

    Enterprise

    Get a customized offer and discounts based on committed use

    Custom
    Contact us
    Full access
    Unlimited gateway requests
    Unlimited GPU hours

    Eligible companies can benefit from our Startup Program. Schedule a call to find out more.