Connect the AI tools your teams already use.
Find current connections for model providers, coding tools, identity systems, infrastructure, and security operations. Every listing states what works today and what remains planned.
- Clear availability labels
- Protocol test tools
- OIDC and AWS support
Providers
Add your OpenAI API key once. Give each app an approved model, a hard spending limit, and a fallback.
BetaAnthropicAdd your Anthropic API key once. Give each app an approved Claude model, a hard spending limit, and a compatible fallback.
AvailableAmazon BedrockPrepay credit, choose a reviewed Bedrock model, and set a workload budget. Managed requests stop when eligible credit cannot cover the required reserve.
BetaAzure OpenAIConnect one Azure OpenAI deployment. Keep its endpoint and key out of application code, then set a separate budget and rate limit for each workload.
BetaGoogle GeminiAdd a Gemini API key once. Give each app an approved model, a hard spending limit, and a compatible fallback.
BetaMistral AIAdd a Mistral API key once. Give each app an approved chat or embedding model, a hard spending limit, and a fallback.
BetaGroqConnect one Groq project. Give each workload an approved model, an immediate spending limit, rate ceilings, and a fallback.
BetaTogether AIConnect one Together AI project. Give each app an approved chat or embedding model, an immediate spending limit, and a compatible fallback.
BetaFireworks AIConnect one Fireworks account. Give each app an approved serverless chat model, an immediate spending limit, rate ceilings, and a compatible fallback.
BetaDeepSeekConnect one DeepSeek API account. Give each app an approved chat model, an immediate spending limit, rate ceilings, and a compatible fallback.
Gateways
Agents & CLIs
Point a separate Codex profile at the gateway. Give the agent a workload key, hard budget, approved model alias, and tested fallback without changing your normal ChatGPT-backed profile.
Client configurationClaude CodeGive Claude Code a separate gateway credential, a hard budget, and an approved Claude route without exposing the provider key.
Client configurationOpenCodeGive OpenCode one workload key and one approved model alias. Keep provider keys off developer machines, cap the workload, and change compatible backends behind the route.