Our take: Pick RunInfra if you want cheap hosted open models behind OpenAI- or Anthropic-compatible endpoints, especially for coding agents with a flat monthly plan.
Main limitation: Pay as you go needs a $5 minimum top-up and has no contractual SLA or SOC 2 Type II; coding plans are capped per 5-hour and weekly windows.
By the AI Tool Atlas editors.
Based on the vendor's official documentation; we have not tested this product hands-on. What “documented overview” means
Is RightNow AI (RunInfra) right for you?
Pick RunInfra if you want cheap hosted open models behind OpenAI- or Anthropic-compatible endpoints, especially for coding agents with a flat monthly plan. Based on the vendor's documentation, not on a hands-on test.
Best for: Developers using coding agents with open models
Also a good fit for
Teams needing low-cost hosted open-model APIs
Consider another tool if $5 minimum top-up for pay as you go
Or if
Pay-as-you-go tier has no contractual SLA and no SOC 2 Type II
Coding plan usage is capped per 5-hour and weekly windows
RightNow AI (RunInfra) pricing and usage limits
Prices checked Oct 2, 2026 on the official US pricing page. Official pricing ↗ Prices and limits change often.
The official RightNow AI (RunInfra) pricing page as we saw it. rightnowai.co, captured Oct 2, 2026
Free access and trial conditions
New accounts start with $1 free to test Model APIs.
Credits beyond plan allowance; images and audio are pay as you go.
Not stated.
What is RightNow AI (RunInfra), and how does it work?
RightNow Research is a YC-backed GPU research lab building tools that close the gap between AI models and the hardware they run on. Its product RunInfra offers GPU infrastructure for deploying, measuring and operating inference workloads, plus hosted Model APIs for open models.
The pricing page lists pay-as-you-go Model APIs from a $5 top-up and a coding plan from $10 a month for tools such as Claude Code, Codex and OpenCode. The lab also publishes papers and open-source projects such as OpenFang and AutoKernel.
RightNow AI (RunInfra)'s official site: how the product presents itself. rightnowai.co, captured Oct 2, 2026
What you provide and what you get back
What you give RightNow AI (RunInfra) and what comes back, from its documentation
You provide
Chat/Messages API requests from SDKs or coding agents
Feature overview on the official site. rightnowai.co, captured Oct 2, 2026
Hosted Model APIs priced per token
Bills hosted open models per million input, cached input and output tokens. Keeps costs predictable for variable workloads.
Access: Pay as you go from a $5 top-up.Limit: Credits last a year per the pricing FAQ.source (opens in a new tab)
OpenAI-compatible chat completions and Anthropic-compatible Messages
Serves chat completions at /v1/chat/completions and Anthropic-style Messages at /v1/messages. Lets existing SDKs and agents work by changing base URL and key.
Coding plan with usage resetting every 5 hours and weekly
Includes a dollar allowance that refreshes every 5 hours and every week, with credits or free Standby at a limit. Gives developers a predictable monthly price for agent use.
Access: Starter $10, Pro $29, Team $99 per month.Limit: Plan covers chat only; images and audio stay pay as you go.source (opens in a new tab)
Cached input discounts on supported models
Discounts repeated prompt prefixes to roughly $0.01 to $0.03 per 1M cached tokens on listed models. Cuts cost of long agent sessions.
Access: Supported models only.Limit: Cache hit rates shown are last-24-hour figures, not guarantees.source (opens in a new tab)
RightNow AI (RunInfra) pros and cons
Where RightNow AI (RunInfra) is strong
Pay as you go with no subscription and no per-seat fees You pay only for tokens used, with nothing for idle capacity.documentedsource (opens in a new tab)
New accounts start with $1 free to test Model APIs Testing costs nothing at first thanks to the starter credit.documentedsource (opens in a new tab)
Compatible with OpenAI and Anthropic API formats Existing OpenAI or Anthropic SDK code works by swapping the endpoint.documentedsource (opens in a new tab)
Limitations to check before paying
$5 minimum top-up for pay as you go Larger compliance needs require an Enterprise contract.documentedsource (opens in a new tab)
Pay-as-you-go tier has no contractual SLA and no SOC 2 Type II Heavy agent use can pause or slow service until a reset.documentedsource (opens in a new tab)
Coding plan usage is capped per 5-hour and weekly windows A cap can stop a long coding session mid-task.documentedsource (opens in a new tab)
How to use RightNow AI (RunInfra) for running coding agents on hosted open models
RightNow AI (RunInfra) documentation: where the steps below come from. rightnowai.co, captured Oct 2, 2026
Set up RightNow AI (RunInfra)
Create a RunInfra workspace and add credits or choose a coding plan.
Create a workspace API key.
Paste the setup prompt into your coding agent or point the SDK at api.runinfra.ai/v1.
Approve the sign-in and the setup plan.
Example: running coding agents on hosted open models
Choose a model such as Qwen3.8 27B.
Send a chat completions request with your key.
Review the response header showing what paid for the request.
Watch usage against the 5-hour limit.
Review and export the result
Confirm the response header shows plan, credits or Standby.
Check the balance after a test run.
Example prompt suggested by our editors (not a recorded output of the product)
Set up RunInfra's open models in this coding agent for me.
RightNow AI (RunInfra) API
RightNow AI (RunInfra) documentation (continued). rightnowai.co, captured Oct 2, 2026
API access and separate billing
RightNow AI (RunInfra) has no public API in its official documentation. OpenAI and Anthropic compatible API documentation ↗
Privacy, data use and commercial rights
Summarized from the vendor's current policies; this is not legal advice and a vendor's badge is not an independent audit.
How your inputs and outputs are handled
Prompts and completions pass through hosted models.
Pricing page states no training on your data.
Not stated in fetched pages.
Commercial use and plan restrictions
SOC 2 Type II and contractual SLA only on Enterprise.
Controls for business users
Workspace isolation
Audit logs and RBAC
TLS in transit, AES-256 at rest
Best RightNow AI (RunInfra) alternatives for different needs
"Better suited" is based on documented features and pricing, not on a head-to-head test.
Not verified: Privacy policy and data retention details not in fetched pages.; Homepage rightnowai.co is a research page; product is RunInfra..
How we documented and maintain this page
Every fact about RightNow AI (RunInfra) above comes from the 5 official pages listed here, last compared on Oct 2, 2026; we have not run RightNow AI (RunInfra) ourselves, so it has no editorial score. How we check and re-check pages.
Article updated Oct 2, 2026; prices checked Oct 2, 2026. Outbound links to vendors are plain links; we do not earn commissions. How we review · Report an error