Releases: radicalbit/radicalbit-ai-gateway
Release list
v0.1.0
v0.1.0 - First Public Release 🎉
This is the first official public release of the Radicalbit AI Gateway!
The feature set is complete, hardened through release candidates, and now ready for production deployments.
What it is
A configurable LLM proxy that sits between your applications and your model providers.
Routing, guardrails, caching, limits, fallback, full request tracing, and MCP proxy — all driven by a
single YAML configuration, managed through a built-in UI, and generatable from plain
English by an AI assistant.
One OpenAI-compatible API in front of every provider you use:
curl http://localhost:9000/v1/chat/completions \
-H "Authorization: Bearer [YOUR_API_KEY]" \
-H "Content-Type: application/json" \
-d '{"model": "[PROJECT_NAME]/[ROUTE_NAME]", "messages": [{"role": "user", "content": "Hello!"}]}'Highlights
🔀 Multi-strategy routing — keyword, token length, context length, time-of-day (cron),
budget ratio, external text classification , and semantic
routing via embedding similarity. Cost control becomes a routing decision.
🛡️ Guardrails in the config, not in the code — pattern matching (contains,
starts_with, ends_with, regex), PII detection and redaction via Microsoft Presidio,
and LLM-as-a-Judge with your own prompt templates. Enforced at the gateway layer, with
block or warn behaviors and per-message-role scoping.
⚡ Caching — exact-match and semantic caching with configurable TTL, similarity
threshold and distance metric, plus exact caching support for audio transcription requests.
🚦 Limits — rate, token and budget limiting per route, independently combinable,
audio-duration-based limiting for transcription routes, and rate limiting for Model Context Protocol (MCP) servers.
🔄 Reliability — automatic model fallback chains that work across providers.
📊 Observability — 20+ Prometheus metrics, end-to-end OpenTelemetry tracing with
ClickHouse storage and custom OTLP exporters (Jaeger, Tempo, …). A usage dashboard with
cost and token breakdown by route, group, and key. Request tagging and tag-based trace
filtering.
🖥️ Built-in UI — projects, configuration editor with load → approve → serve lifecycle,
config import/export, groups and API keys, usage dashboard and trace inspection.
🤖 Configure in plain English — describe the route you want and get valid YAML, either
from the Generate button in the UI or from your IDE via the Radicalbit Skills plugin.
🚨 Real-time Alerts & Notifications — proactive email notifications triggered on critical
gateway events: guardrail interventions (input/output violations, PII detected/redacted, LLM judge breaches). Rules are
configurable per route with instant delivery and customizable recipient lists.
🔌 Providers — native support for OpenAI, Anthropic, Google Gemini, DeepSeek, Mistral
and Azure OpenAI, plus any OpenAI-compatible endpoint (Ollama, vLLM, OpenRouter,
on-premises deployments).
🔌 MCP proxy — expose Model Context Protocol servers through the same route,
the same API key with the same trace view as your models. One endpoint per route
fans out across several upstreams over Streamable HTTP or stdio, namespacing
tools as {alias}__{tool}.
Notes
- The commit history starts at the repository bootstrap, not at the beginning of development.
Contributors
- @rivamarco made their first contribution in #1
- @dtria91 made their first contribution in #3
- @lucataglia made their first contribution in #5
- @carlo-pignatiello made their first contribution in #9
- @gipster made their first contribution in #12
- @bigmoby made their first contribution in #29
- @mmariniello90 made their first contribution in #8
- @maocorte made their first contribution in #73
Full Changelog: https://github.com/radicalbit/radicalbit-ai-gateway/commits/v0.1.0
Contributors
Assets 2
v0.1.0-rc.3
v0.1.0-rc.3
It is a release candidate, not a stable release: the feature set is complete and running
internally, and the tag exists so it can be pinned, deployed and tested as a fixed
version before we cut v0.1.0.
What it is
A configurable LLM proxy that sits between your applications and your model providers.
Routing, guardrails, caching, limits, fallback and full request tracing — all driven by a
single YAML configuration, managed through a built-in UI, and generatable from plain
English by an AI assistant.
One OpenAI-compatible API in front of every provider you use:
curl http://localhost:9000/v1/chat/completions \
-H "Authorization: Bearer YOUR_API_KEY" \
-H "Content-Type: application/json" \
-d '{"model": "quickstart/my-assistant", "messages": [{"role": "user", "content": "Hello!"}]}'Highlights
🔀 Multi-strategy routing — keyword, token length, context length, time-of-day (cron),
budget ratio, external text classification, LLM-based intent classification, and semantic
routing via embedding similarity. Cost control becomes a routing decision.
🛡️ Guardrails in the config, not in the code — pattern matching (contains,
starts_with, ends_with, regex), PII detection and redaction via Microsoft Presidio,
and LLM-as-a-Judge with your own prompt templates. Enforced at the gateway layer, with
block or warn behaviors and per-message-role scoping.
⚡ Caching — exact-match and semantic caching with configurable TTL, similarity
threshold and distance metric.
🚦 Limits — rate, token and budget limiting per route, independently combinable,
plus audio-duration-based limiting for transcription routes.
🔄 Reliability — automatic model fallback chains that work across providers.
📊 Observability — 20+ Prometheus metrics, end-to-end OpenTelemetry tracing with
ClickHouse storage and custom OTLP exporters (Jaeger, Tempo, …), a usage dashboard with
cost and token breakdown by route and group, request tagging and tag-based trace
filtering, and email alert rules on guardrail, caching and route events.
🖥️ Built-in UI — projects, configuration editor with load → approve → serve lifecycle,
config import/export, groups and API keys, usage dashboard and trace inspection.
🤖 Configure in plain English — describe the route you want and get valid YAML, either
from the Generate button in the UI or from your IDE via the Radicalbit Skills plugin
for Claude Code.
🔌 Providers — native support for OpenAI, Anthropic, Google Gemini, DeepSeek, Mistral
and Azure OpenAI, plus any OpenAI-compatible endpoint (Ollama, vLLM, OpenRouter,
on-premises deployments).
Notes
- This is a prerelease — APIs and configuration schema may still change before
v0.1.0. - The commit history starts at the repository bootstrap, not at the beginning of
development.
What's Changed
- chore: fix readme and compose by @rivamarco in #78
- add images for the readme file by @mmariniello90 in #80
- fix: generator prompt and readme by @carlo-pignatiello in #79
- feat: add exact caching support for transcription requests by @dtria91 in #77
Full Changelog: v0.1.0-rc.2...v0.1.0-rc.3
Contributors
Assets 2
v0.1.0-rc.2
v0.1.0-rc.2 — first prerelease 🎉
This is the first prerelease of the Radicalbit AI Gateway, so this changelog covers
the entire history of the project to date.
It is a release candidate, not a stable release: the feature set is complete and running
internally, and the tag exists so it can be pinned, deployed and tested as a fixed
version before we cut v0.1.0.
What it is
A configurable LLM proxy that sits between your applications and your model providers.
Routing, guardrails, caching, limits, fallback and full request tracing — all driven by a
single YAML configuration, managed through a built-in UI, and generatable from plain
English by an AI assistant.
One OpenAI-compatible API in front of every provider you use:
curl http://localhost:9000/v1/chat/completions \
-H "Authorization: Bearer YOUR_API_KEY" \
-H "Content-Type: application/json" \
-d '{"model": "quickstart/my-assistant", "messages": [{"role": "user", "content": "Hello!"}]}'Highlights
🔀 Multi-strategy routing — keyword, token length, context length, time-of-day (cron),
budget ratio, external text classification, LLM-based intent classification, and semantic
routing via embedding similarity. Cost control becomes a routing decision.
🛡️ Guardrails in the config, not in the code — pattern matching (contains,
starts_with, ends_with, regex), PII detection and redaction via Microsoft Presidio,
and LLM-as-a-Judge with your own prompt templates. Enforced at the gateway layer, with
block or warn behaviors and per-message-role scoping.
⚡ Caching — exact-match and semantic caching with configurable TTL, similarity
threshold and distance metric.
🚦 Limits — rate, token and budget limiting per route, independently combinable,
plus audio-duration-based limiting for transcription routes.
🔄 Reliability — automatic model fallback chains that work across providers.
📊 Observability — 20+ Prometheus metrics, end-to-end OpenTelemetry tracing with
ClickHouse storage and custom OTLP exporters (Jaeger, Tempo, …), a usage dashboard with
cost and token breakdown by route and group, request tagging and tag-based trace
filtering, and email alert rules on guardrail, caching and route events.
🖥️ Built-in UI — projects, configuration editor with load → approve → serve lifecycle,
config import/export, groups and API keys, usage dashboard and trace inspection.
🤖 Configure in plain English — describe the route you want and get valid YAML, either
from the Generate button in the UI or from your IDE via the Radicalbit Skills plugin
for Claude Code.
🔌 Providers — native support for OpenAI, Anthropic, Google Gemini, DeepSeek, Mistral
and Azure OpenAI, plus any OpenAI-compatible endpoint (Ollama, vLLM, OpenRouter,
on-premises deployments).
Notes
- This is a prerelease — APIs and configuration schema may still change before
v0.1.0. - The commit history starts at the repository bootstrap, not at the beginning of
development.
New Contributors
- @rivamarco made their first contribution in #1
- @dtria91 made their first contribution in #3
- @lucataglia made their first contribution in #5
- @carlo-pignatiello made their first contribution in #9
- @gipster made their first contribution in #12
- @bigmoby made their first contribution in #29
- @mmariniello90 made their first contribution in #8
- @maocorte made their first contribution in #73
Full Changelog: https://github.com/radicalbit/radicalbit-ai-gateway/commits/v0.1.0-rc.2