What is AxonHub?
AxonHub is an open-source AI gateway for developers and teams connecting applications to multiple model providers. AxonHub translates requests into formats compatible with supported providers, allowing an application to retain its existing SDK while changing its provider configuration. The repository describes support for OpenAI, Anthropic and other AI SDKs.
AxonHub addresses provider switching, integration complexity, request visibility and cost management within the same gateway. AxonHub includes an administrative interface with dashboards, channel management, model pricing, trace viewing and request monitoring. The project is distributed through a public repository, with documentation covering API references, architecture, development and deployment.
What AxonHub can do
AxonHub combines API translation with controls for routing, access, observability and spending. AxonHub supports text generation, image generation, embeddings and reranking, although the provider table lists different API capabilities for different providers. Supported providers include OpenAI, Anthropic, Gemini, DeepSeek, Zhipu, Moonshot, OpenRouter, Jina AI and NanoGPT.
- AxonHub translates requests between SDK formats and supported model providers.
- AxonHub routes requests with load balancing and automatic failover.
- AxonHub records complete request timelines with thread-level tracing.
- AxonHub tracks input, output and cached token costs for each request.
- AxonHub provides role-based access control, usage quotas and data isolation.
- AxonHub supports budget management and real-time usage tracking.
How AxonHub works
AxonHub documents deployment through instructions that an agent can read and follow. After deployment, an administrator opens the returned access address and creates an administrator account through the initialization wizard. AxonHub then requires provider channels to be added with provider API keys, supported models configured, connections tested and channels enabled.
AxonHub issues a separate API key for client applications through the API Keys interface. The documented OpenAI example points the SDK base URL at the AxonHub endpoint and supplies that gateway key. AxonHub handles the provider-facing request translation, while the application continues using its existing SDK request structure.
Who AxonHub is for
AxonHub serves developers integrating AI models and teams managing access to shared model providers. AxonHub is relevant when applications need provider flexibility without SDK replacement, or when teams need request traces, spending visibility and access controls around model calls. The documented API types also cover applications using images, vector embeddings or reranked results.
- AxonHub supports applications that switch between model providers through configuration.
- AxonHub helps teams inspect request histories when investigating integration problems.
- AxonHub supports shared model access with quotas and role-based permissions.
- AxonHub tracks request costs for teams managing model usage and budgets.
AxonHub pricing and access
AxonHub is available as an open-source project under Apache-2.0 and LGPL-3.0 licences, with the scope of each licence described in NOTICE. AxonHub also provides a demonstration instance configured with free Zhipu and OpenRouter models. The repository links to Bloome as an option for running AxonHub through a browser or phone without local environment configuration.
AxonHub strengths and limitations
AxonHub brings provider translation, routing, request tracing and cost tracking into a shared gateway. AxonHub also documents an initialization workflow and provides a demonstration instance for evaluation. Capability coverage is not uniform across the provider table, and several functions remain unfinished or under testing, so deployment planning should account for the documented support status.
- AxonHub lists realtime conversation functionality as a planned API type.
- AxonHub marks AWS Bedrock and Google Cloud support as being tested.
- AxonHub requires provider credentials and channel setup for a configured deployment.
AxonHub FAQ
What is AxonHub?
AxonHub is an open-source AI gateway that translates SDK requests into formats compatible with supported model providers. AxonHub combines provider switching with load balancing, automatic failover, request tracing, access controls and cost tracking for developers and teams building AI applications.
How is AxonHub available?
AxonHub is available through a public repository under Apache-2.0 and LGPL-3.0 licences, with licensing scope explained in NOTICE. AxonHub also offers a demonstration instance configured with free models from Zhipu and OpenRouter.
Can AxonHub use the OpenAI SDK with other model providers?
AxonHub supports using the OpenAI SDK with supported providers such as Anthropic through request translation. AxonHub documents configuring the SDK with the gateway base URL and an AxonHub API key, allowing provider changes through configuration rather than application code changes.
What types of AI requests does AxonHub support?
AxonHub supports text generation, image generation, embeddings and reranking. AxonHub lists compatible APIs by provider, so available request types vary across providers. Realtime conversation functionality is listed as planned rather than completed.
How does AxonHub track model costs and usage?
AxonHub provides real-time usage tracking, budget management and per-request cost details covering input, output and cached tokens. AxonHub also includes usage quotas and fine-grained access controls for managing model access across teams.
From the community
No reviews yet
Be the first to review this tool. What you write helps the next person choose.




