AISBF
AI Service Broker Framework — AI Should Be Free
AISBFAI Service Broker Framework — AI Should Be Free
Concrete guides for using AISBF as an AI gateway, routing layer, Studio/BFInX proxy, cluster-aware broker, and production control plane — plus CoderAI tutorials for running local multimodal AI workers with or without AISBF.
Install the BYOK/BYOT Chrome extension and route browser prompts through hosted or self-hosted AISBF.
One detailed walkthrough covering install, dashboard usage, network graph, clusters, rotation/autoselect concepts, and CoderAI+RunPod coupling.
CoderAI works on its own: run local multimodal workers, use the browser UI, or call its OpenAI-compatible API directly. AISBF integration is optional when you want routed, metered, clustered access.
Explore CoderAI →
Read CoderAI documentation →
Install CoderAI standalone →
Use the CoderAI API →
Optional AISBF broker tutorial →
See the three bundled demo apps that use CoderAI as a local multimodal backend: the video editor, VideoGen Studio, and Township Fighters generator.
View tools overview →
Read demo tools docs →
Run them in Docker / OCI →
Standalone dark diagram explaining production and clustered AISBF topology.
Step-by-step service user, MySQL, Redis, systemd, nginx, TLS edge, and verification.
Route your app through AISBF instead of wiring every provider directly.
Use rotations, fallbacks, quotas, and route separation.
Keep sensitive prompts on local or self-hosted backends.
Add OpenAI-compatible, OAuth, Ollama, RunPod, or CoderAI providers.
Use rotations for reliability and autoselect for policy.
Call image, audio, video, embedding, and advanced endpoints through AISBF.
Understand provider, rotation, and policy route boundaries before wiring production traffic.
Optional integration: use AISBF as the stable control plane above standalone CoderAI workers and burst GPU endpoints.
Configure shared MySQL, Redis, response cache, MCP tokens, and load balancing.
Daily dashboard checks, safe route changes, cache operations, and external smoke tests.
Concrete patterns for privacy-first routing, cost-aware coding routes, RunPod bursts, and model-family separation.
Diagnose multi-node drift across load balancers, MySQL, Redis, response cache, tokens, and route smoke tests.
OpenAI-compatible APIs, provider routes, rotations, autoselect, user-scoped endpoints, dashboard, and MCP.