# bricks-cloud/BricksLLM

🔒 Enterprise-grade API gateway that helps you monitor and impose cost or rate limits per API key. Get fine-grained access control and monitoring per user, application, or environment. Supports OpenAI, Azure OpenAI, Anthropic, vLLM, and open-source LLMs.

Repository: https://github.com/bricks-cloud/BricksLLM
Canonical: https://ross.abutalabs.com/products/bricksllm
Homepage: https://trybricks.ai/
Language: Go
License: MIT
License Family: permissive
Topics: golang, llm, openai, ai, anthropic, azure, gpt, postgresql, rest-api, ycombinator, api, docker, privacy, security, artificial-intelligence, generative-ai, open-source, self-hosted, vllm
Last push: 2025-01-05T23:47:45+00:00

## Health v2 (maintenance only)
Score: 30/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 0, release rhythm 40, longevity 81
- inputs: {"age_days": 1142, "days_push": 605, "days_rel": 605, "gap_med": 1.0, "n_releases_24m": 19}
- flags: none
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 1226, forks 97 (observed 2026-08-28T04:04:03.220228+00:00)

## What it is
BricksLLM is a cloud-native AI gateway written in Go that sits between applications and LLM providers like OpenAI, Anthropic, Azure OpenAI, and vLLM. It provides per-API-key rate limiting, cost control, usage analytics, PII detection and masking, caching, retries, and failover for putting LLMs into production.

## Use cases
- set llm usage and cost limits per api key
- track llm usage per user and organization
- redact pii from llm requests
- add failover and retries between openai and azure openai
- rate limit llm api keys for students or pricing tiers
- self-host an llm api gateway with monitoring
- cache llm responses to cut costs

## When to choose
- you need enterprise-grade per-key rate and cost limits for LLM traffic
- you want self-hosted control over LLM gateway privacy and PII masking
- you need failover, retries, and caching across multiple LLM providers
- you want per-user or per-environment LLM usage analytics

## When to avoid
- you only need a simple proxy without rate limiting or cost control
- you want a fully managed SaaS gateway with a dashboard out of the box
- your stack requires providers not natively supported and you cannot write custom integrations

## Facets
- artifact type: service
- maturity: active
- function: api-gateway, rate-limiting, caching, monitoring, auth, middleware, proxy
- domain: large-language-models, artificial-intelligence, apis, security, self-hosted, analytics
- platform: self-hosted, go, cloud
- tags: llm-gateway, cost-control, pii-detection, openai-proxy, anthropic, azure-openai, vllm, api-key-management, failover, docker, linux

## Member repositories
- bricks-cloud/BricksLLM (main) score 30

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:04:03.220228+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-30T06:15:18.347812+00:00, confidence not recorded.
  - readme: https://github.com/bricks-cloud/BricksLLM (fetched 2026-08-28T04:04:03.220228+00:00, sha a35b0000784d)
  - homepage: https://trybricks.ai/ (fetched 2026-08-29T12:23:09.874527+00:00, sha 1c813d508ec5)
  - site_page: https://whitecircle.com/?scrollTo=features (fetched 2026-08-29T12:23:09.883981+00:00, sha a7a32d923884)
  - site_page: https://docs.whitecircle.com (fetched 2026-08-29T12:23:09.886166+00:00, sha 61cab858bf3c)
- Data as of 2026-08-30T08:39:29.467469+00:00.