Our take: Choose AcruxCore if your engineering team wants prompts, an OpenAI-compatible gateway, tracing and evaluation in one open-source platform that you can also self-host.
Main limitation: Hosted pricing after the beta has not been set; the vendor says it will publish it on the pricing page.
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
Who AcruxCore fits
Choose AcruxCore if your engineering team wants prompts, an OpenAI-compatible gateway, tracing and evaluation in one open-source platform that you can also self-host. Based on the vendor's documentation, not on a hands-on test.
Best for: Engineering teams building LLM features
Also a good fit for
Teams needing self-hosted LLM observability
Consider another tool if post-beta pricing not set
Or if
Requires own provider API keys
Account required for hosted version
AcruxCore pricing and usage limits
Prices checked Sep 29, 2026 on the official US pricing page. Official pricing ↗ Prices and limits change often.
The official AcruxCore pricing page as we saw it. acruxcore.com, captured Oct 1, 2026
Free access and trial conditions
Hosted platform is free during beta; self-hosting is free under Apache-2.0.
No separate trial; beta needs no credit card at sign-up.
Model tokens are billed by your provider at their rates; AcruxCore adds no markup.
Not stated on the pricing page.
What is AcruxCore, and how does it work?
AcruxCore is an open-source, self-hostable LLMOps platform for engineering teams. It lets teams version prompts, route LLM calls through an OpenAI-compatible gateway, trace requests, catalog tools and run evaluations.
The hosted platform is free during beta; users bring their own provider keys and pay providers directly.
AcruxCore's official site: how the product presents itself. acruxcore.com, captured Oct 1, 2026
What you provide and what you get back
What you give AcruxCore and what comes back, from its documentation
You provide
Prompt templates and tool definitions
LLM calls sent through the gateway
OpenTelemetry spans from your own agents
Evaluation datasets
You get back
Versioned prompts with movable aliases
Traces with model, tokens, latency and cost
Evaluation scores and optimizer-drafted prompt rewrites
Audit trail of changes
Where you can use it
Platforms
Web, API
Account
Required
Licence
Apache-2.0, no enterprise-only directory
SDKs
TypeScript and Python
Deployment
Hosted or self-hosted via docker compose
Providers
OpenAI, Anthropic, Gemini, OpenAI-compatible
Best use cases for AcruxCore
Scenarios are worked out by our editors from the documentation; the actual result depends on your input and plan.
Versioning prompts without redeploying
Who
Engineers shipping LLM features
You provide
A templated prompt and variables
Expected result
A rendered prompt served by alias, switched between versions without an app deploy
Accepts OpenAI-style requests and forwards them to connected OpenAI, Anthropic, Gemini or compatible providers. Existing OpenAI clients work after changing the base URL.
Not verified: Post-beta hosted pricing; Trace retention windows.
How we documented and maintain this page
Every fact about AcruxCore above comes from the 3 official pages listed here, last compared on Sep 29, 2026; we have not run AcruxCore ourselves, so it has no editorial score. How we check and re-check pages.
Article updated Oct 10, 2026; prices checked Sep 29, 2026. Outbound links to vendors are plain links; we do not earn commissions. How we review · Report an error
(open full size in a new tab)
(open full size in a new tab)
(open full size in a new tab)
(open full size in a new tab)
(open full size in a new tab)
(open full size in a new tab)