bitsrun

Pay per second while your model runs

No subscription required. Buy credits, deploy, and the meter only runs from the moment your model is ready until your idle timeout ends.

Private endpoints

  • NVIDIA L4, 24 GB

    7 to 14B quantized LLMs, any embedding model

    Per hour
    $0.95
    Per second
    $0.000264
    Always on
    $1,180/mo

  • NVIDIA L40S, 48 GB

    14 to 32B quantized, 8 to 14B full precision

    Per hour
    $3.20
    Per second
    $0.000889
    Always on
    $3,974/mo

  • 4x NVIDIA L4, 96 GB

    Full-precision 27 to 32B, 70B at 4-bit

    Per hour
    $8.52
    Per second
    $0.002367
    Always on
    $6,843/mo

Not sure which size you need? Paste any model into the model sizer and it shows what fits and what it costs. Smaller slices of an L4 (3 GB and 6 GB) for embeddings and small models are coming.

Plans

Every plan uses the same prices. Pro and Business include credits each month, more endpoints and bonus credit on top-ups.

  • Pay as you go

    $0 / month

    Prepaid, top up from $10

    • 2 private endpoints
    • 10 requests at once
    • 10 GB free model storage
    • Docs and email support
    Get started

  • Pro

    $29 / month

    $29 of credits every month

    • 10 private endpoints
    • 50 requests at once
    • 5% bonus on top-ups
    • 200 GB free model storage
    • Email support, reply target 2 business days
    Choose Pro

  • Business

    $299 / month

    $299 of credits every month

    • 50 private endpoints
    • 150 requests at once
    • 10% bonus on top-ups
    • 1 TB free model storage
    • First in line for cold starts
    • Email and shared Slack, reply target 1 business day
    Choose Business

Reply times are targets, not guarantees.

Good to know

  • Minimum top-up is $10. Credit you pay for does not expire.
  • Priority is 1.7 times the rate: a cold start skips the wait for spare GPUs and launches on a dedicated one right away.
  • Model storage is free up to 10 GB on Pay as you go, 200 GB on Pro and 1 TB on Business, then $0.05 per GB each month.
  • If you think a charge is wrong, email support@bitsrun.com from the address you sign in with. The Refund Policy explains what we refund.

An example month

A chatbot on an L4 that is busy 3 hours a day, including its idle timeout:

$85.54

90 hours at $0.95 per hour. The same GPU running all month would cost $693.79.