AISBFAI routing, built in public.
AISBF is the control plane for practical AI work: route requests across cloud providers, local CoderAI workers, rotations, failover paths and privacy rules from one OpenAI-compatible gateway.
CoderAI: orchestrate your models.
Turn your own GPU machine into a local multimodal worker for chat, images, audio, video and private workloads — while AISBF handles routing, failover and cloud/local policy.
Testing-period support plan: €6/month or €60/year for unlimited Pro access while AISBF is in active development.
install aisbf
Install from PyPI and start the local AISBF server.
python3 -m pip install --upgrade aisbf
aisbf
# opens http://127.0.0.1:17765/dashboardinstall coderai
Download the CoderAI Docker install bundle, extract it, then run the bundled provider installer.
curl -L -o coderai-docker-dist.tar https://aisbf.cloud/coderai-docker-dist.tar
tar -xf coderai-docker-dist.tar
cd coderai-docker-dist
./install.sh
coderai-docker --nvidia --detachWhat AISBF gives you
Lead with useful infrastructure, then make the support model clear. Transparency, not apology.
Useful now
A real hosted demo for routing across providers, models, local backends and failover paths.
Still early
Clear demo-stage expectations from the first screen while the platform is actively developed.
Founder-led
The one-human nature is part of the story: human, credible, and transparent.
Support = access
€6/month or €60/year is both unlimited hosted Pro during testing and a way to keep AISBF moving.
Choose your path
Use the hosted demo now, self-host when control matters, or start with the specific angle that fits your work.
Early adopter support
Best if you want unlimited Pro during testing and want to help fund continued development.
For developers
One gateway for multi-provider AI, cleaner routing, and less provider glue code.
Privacy & self-host
Local-first routing, TOR-friendly exposure, self-hosting, and less vendor lock-in.
Browser companion
A quick browser popup that talks to hosted or self-hosted AISBF routes.
Teams, ops & admin controls
Quotas, analytics, multi-user operations, and governance over model/provider access.
CoderAI: local multimodal workers for AISBF
AISBF is the routing and control plane. CoderAI is the local compute side: a machine you control can expose OpenAI-compatible and multimodal generation endpoints, then register with AISBF as routable provider capacity.
Your GPU, your worker
Run CoderAI on a workstation, home-lab box, RunPod instance, or cloud GPU server and make it available through AISBF routing instead of wiring apps directly to that machine.
More than chat completions
CoderAI covers local chat, embeddings, images, video, TTS, STT, voice cloning, face swap, 3D conversion, archives, and custom generation pipelines.
Outbound broker mode
If the worker sits behind NAT or a private network, CoderAI can dial AISBF over WSS and receive brokered work without exposing an inbound public port.
Local + cloud policy
Use AISBF rules to decide what goes to CoderAI, what goes to cloud providers, and what fails over when a provider or worker is unavailable.
AISBF Market: sell access to your models
Turn provider capacity into something people can actually use: controlled routes, quotas, broker sessions, and OpenAI-compatible access instead of raw endpoint sharing.
For model owners
Publish access to local models, CoderAI workers, private fine-tunes, RunPod boxes, or other OpenAI-compatible providers through AISBF-controlled routes.
For local GPU sellers
Use CoderAI broker mode to connect outward over WSS, advertise capabilities, and receive routed work without exposing your machine directly to the public internet.
For buyers and apps
Call stable AISBF routes while the platform handles provider identity, routing policy, failover, quotas, and cloud/local selection.
Built for real AI operations
AISBF is not just a thin proxy. It is a control layer for routing, resilience, cost, privacy, and operational visibility.
Prompt & context-aware routing
Analyze prompts and context before routing so work can land on the provider or model that fits the task.
Multi-provider orchestration
Unify Google, OpenAI, Anthropic, Ollama, and other backends behind one interface.
Security & privacy controls
Self-host, route to local models, restrict exposure, and control where model traffic goes.
Context condensation
Condense long prompts and conversations with multiple strategies to stay within model limits.
Resilience & failover
Use weighted balancing, fallback routing, and provider failover when a backend slows down or disappears.
Rate limits, quotas & analytics
Track usage, enforce TPM/TPH/TPD controls, and understand spend by user, provider, and model.
Multi-user operations
Run AISBF for teams, customers, or projects with per-user keys, quotas, and activity visibility.
Streaming & real-time APIs
Support interactive workloads with streaming responses and low-latency delivery.
TOR & local-first deployment
Expose AISBF through TOR, keep sensitive workloads local, and reduce reliance on any single vendor.
Hosted now, self-host later
AISBF stays open source either way. The hosted path is about speed, convenience, and supporting continued development.
Use hosted AISBF
Best when you want results now instead of another deployment project.
- Instant dashboard access
- No server setup or maintenance
- Fast way to test the full platform
- Unlimited Pro during testing for €6/month or €60/year
Self-host AISBF
Best when you want maximum control over infrastructure, privacy, and routing behavior.
- Run it on your own hardware or servers
- Route to local models and private backends
- Audit the code and operate it your way
- Good for privacy-sensitive deployments
Practical tutorials
Copyable guides for deploying AISBF, routing across providers, adding failover, and operating the dashboard.
Install AISBF in Production
Service user, MySQL, Redis, systemd, nginx, TLS, and verification.
Build a Multi-Provider AI Gateway
Route your app through AISBF instead of wiring every provider directly.
Add Provider Failover and Smart Routing
Use rotations, fallback routes, quotas, and route separation for reliability.
Route Private Workloads to Local Models
Keep sensitive prompts on local or self-hosted backends.
Create Providers, Rotations and Autoselect
Use the dashboard to build failover rotations and autoselect routes.
BFInX / Studio Proxy and Cluster Setup
Proxy media and embedding workloads and connect broker workers.
Full API Reference
OpenAI-compatible APIs, user-scoped routes, rotations, autoselect, dashboard surfaces, and MCP.
CoderAI and RunPod Integration
Use AISBF above local broker workers and burst GPU endpoints.
Dashboard Operations Runbook
Provider checks, route changes, cache validation, and smoke tests.
Advanced Routing and Cluster Troubleshooting
Privacy-first, cost-aware, burst GPU routing recipes and multi-node debugging.
Cluster Troubleshooting Playbook
Node-by-node checks and route smoke-test matrices.
Simple pricing
Free/open source if you self-host. €6/month or €60/year during testing if you want the hosted service and want to support AISBF.
Free
Best for personal testing, evaluation, and small self-hosted workflows.
- Single-user access
- Core routing and failover
- Basic quotas and rate limits
- Community support
Pro Test Plan
Unlimited hosted usage during the testing period for early adopters who want the full platform and want to support development.
- Unlimited requests during testing
- Unlimited providers, rotations, and autoselections
- Early-adopter access while AISBF evolves
- Temporary pricing that may change after testing
Self-Hosted
Run AISBF your own way with maximum control over privacy, providers, and infrastructure.
- Open source deployment
- Local + cloud hybrid support
- TOR / privacy-conscious exposure
- Operator-controlled architecture
AISBF vs LiteLLM
Both are capable OSS tools. AISBF is more opinionated about prompt-aware routing, context handling, local-cloud mixing, and privacy-conscious deployment.