> ## Documentation Index
> Fetch the complete documentation index at: https://runinfra.ai/docs/llms.txt
> Use this file to discover all available pages before exploring further.
# Plans and pricing
> Compare the Core (self-serve, custom $50-$1000/mo) and Enterprise plans, the unified credit balance, free starting credits, and token pricing.
RunInfra has two plans on one unified credit balance: **Core**, a self-serve plan where you choose a custom monthly amount, and **Enterprise** for custom volume and dedicated infrastructure. Usage for the agent, optimization runs, benchmarking, deploys, and inference all draws from the same balance, where **1 credit = \$1**.
New accounts start with **\$10 in free credits**. No credit card required to try the platform.
## Plans
| Feature | Core | Enterprise |
| ---------------------------------- | ----------------------------------- | ----------------------------- |
| **Price** | Custom $50-$1000 / month | Custom (above \$1000/mo) |
| **Credits** | Your monthly amount, 1 credit = \$1 | Custom credit volume |
| **Credit balance** | One unified balance | One unified balance |
| **Seats** | No per-seat fees | No per-seat fees |
| **Optimization** | AWQ, GPTQ, FP8 quantization | Everything in Core |
| **Standard GPUs (T4-H100)** | Yes | Yes |
| **B200 / H200 GPUs** | No | Yes |
| **Managed deploy + scale-to-zero** | Yes | Yes |
| **OpenAI-compatible API** | Yes | Yes |
| **Self-hosted / custom GPU** | No | Yes |
| **Audit logs + RBAC** | RBAC | Audit logs + RBAC |
| **SLA** | 99.9% | Custom, up to 99.99% |
| **Compliance** | - | SOC 2 Type II |
| **Support** | Priority email | Dedicated CSM + private Slack |
**Core** is self-serve at a custom monthly amount you choose between **$50 and $1000**. Your payment converts to credits (1 credit = \$1) in a single unified balance that funds optimization runs, the agent, benchmarking, deploys, and inference alike. There are no seats and no per-user fees. Core includes quantization (AWQ, GPTQ, FP8), all standard GPUs (T4, L4, L40S, A100, H100), managed deploy with scale-to-zero endpoints, OpenAI-compatible API endpoints, and unlimited pipelines and versioning.
**Enterprise** is for monthly volume above \$1000 and adds self-hosted and custom-GPU deployment, B200 / H200 GPU access, audit logs, custom SLAs up to 99.99%, SOC 2 Type II compliance, a custom credit volume with contract terms, and a dedicated CSM with a private Slack channel. [Contact sales](https://runinfra.ai/contact) for a quote.
## Token pricing
When you deploy an endpoint, inference is billed per million tokens, drawn from your unified credit balance. The table below shows estimated starting rates by model size; your actual cost depends on your pipeline configuration.
| Model size | Input (from) | Output (from) |
| -------------- | ------------- | ------------- |
| Small (1-8B) | \$0.08 / MTok | \$0.20 / MTok |
| Medium (8-30B) | \$0.20 / MTok | \$0.80 / MTok |
| Large (30-70B) | \$0.45 / MTok | \$1.50 / MTok |
| XL (70B+) | \$0.80 / MTok | \$2.50 / MTok |
Your actual per-token cost depends on your full pipeline: model choice, quantization method, GPU tier, routing, and deployment mode. RunInfra shows the projected cost in the deploy tab before you go live, derived from your real benchmark measurements rather than catalog defaults.
Volume pricing is an Enterprise conversation. [Contact sales](https://runinfra.ai/contact) to discuss custom monthly volume above \$1000/mo.
## One unified credit balance
Everything draws from a single credit balance where **1 credit = \$1**. There is no split between "optimization" and "inference" credits, the agent, optimization runs, benchmarking, deploys, and production inference all debit the same balance.
| What it funds | How it's billed |
| -------------------------------------------------- | --------------------------------------------------------------------------------------------------- |
| Agent chat, plans, optimization runs, benchmarking | Metered to the measured GPU cost of each run, drawn from your balance |
| Production endpoint traffic | Per-million-tokens (Flex scale-to-zero and Active always-on); Active adds a reserved-GPU hourly fee |
New accounts start with \$10 in free credits. Add credits any time at [Settings > Cost](https://runinfra.ai/settings/cost#credits), where you can also track your balance and activity.
## What counts as usage
One optimization run on one pipeline, a single call to start optimization, regardless of how many model variants are benchmarked in that run. When a run starts, RunInfra places a temporary hold on your credit balance and settles it to the measured GPU cost once the run finishes, refunding the unused amount. Failed or cancelled runs are refunded in full.
Input tokens (your prompt) and output tokens (the model's response) are counted and billed separately at the rates shown in the token pricing table above, drawn from your unified credit balance.
Credits are your single prepaid balance (1 credit = $1) used for everything: the agent, optimization, benchmarking, deploys, and inference. New accounts get $10 in free credits. Credits do not expire while your account is active.
Active (always-on) endpoints keep a GPU warm 24/7 for zero cold start. The reserved GPU time is metered against your unified credit balance; the Deploy tab shows the projected monthly cost before you go live.
## Manage your plan
Choose or adjust your Core monthly amount and view billing history at [Settings > Billing](https://runinfra.ai/settings/billing); add credits and track your balance at [Settings > Cost](https://runinfra.ai/settings/cost#credits). To move above \$1000/mo, [contact sales](https://runinfra.ai/contact) about Enterprise.
## Next steps
Create an account and deploy your first endpoint in 5 minutes.
How GPU selection and deployment mode affect per-token cost.
Flex scale-to-zero and Active always-on endpoints.
Answers to the most common questions about RunInfra.