Groq API and Dome
Fast open-weight models on Groq, with every call governed. Groq runs open-weight models like gpt-oss-120b and Llama. Connect it to Dome and a rule can hold an agent to open weights, whatever model name it asks for. OpenRouter answers for gpt-oss-120b when Groq can't.
Dome provides model brokeringand routing for Groq.
Groq's key, Dome's custody
The Groq key is stored in Dome and attached per request. An agent only ever has its Dome key.
Groq first, OpenRouter second
Both serve gpt-oss-120b. In one pool, OpenRouter answers whatever Groq drops.
Open weights by rule
Tag connections that serve open-weight models. A rule keeps an agent on them, whatever model name it sends.
Connect Groqin three steps.
Add gpt-oss-120b on Groq, put OpenRouter behind it, and aim agents at the Gateway.
Add the connection
The provider id is groq. Dome uses https://api.groq.com/openai/v1 and Groq's own model ids.
shelldome models add gpt-oss-groq \ --provider groq \ --model openai/gpt-oss-120b \ --api-key "$GROQ_API_KEY" \ --attributes '{"weights":"open"}' \ --gateway prod-gatewayPool it with OpenRouter
Groq is priority 0, and agents ask for the pool's name, gpt-oss-120b.
shelldome models pool create gpt-oss-120b \ --failover-max all --gateway prod-gateway dome models pool member add gpt-oss-120b gpt-oss-groq --priority 0dome models pool member add gpt-oss-120b gpt-oss-openrouter --priority 1Point your agent at the Gateway
Any client that speaks the OpenAI API works. Swap in the Gateway URL and a Dome agent key.
pythonfrom openai import OpenAI client = OpenAI(base_url=f"{GATEWAY_URL}/v1", api_key=DOME_AGENT_KEY)client.chat.completions.create( model="gpt-oss-120b", messages=[{"role": "user", "content": "Route this message: sales, support or billing."}],)
This agent usesopen-weight models only.
Applied to one agent with --agent, this refuses any connection not tagged weights: open. The intent router can't reach a frontier model by asking for one.
cedar
forbid (principal, action == Dome::Action::"llm:invoke", resource is Dome::LLMModel)unless { resource has weights && resource.weights == "open" };Bringing it alltogether.
Connecting Groq to registered agents, tools, and identity in Dome completes a governed agent application.
Acting for
Agent
Control point
GatewayEvery call decided and audited
RulesGuardsQuotas
Models
GroqProvider
OpenRouterProviderCommon questionsabout Groq.
More of whatDome works with.
Model
Claude Fable
Fable 5.1 from Anthropic and Amazon Bedrock in one failover pool, open to one group and capped by quota.
MCP server
GitHub
The GitHub MCP server behind the Tool Gateway, with rules that decide each call on its owner and repository.
Identity
Okta
Okta tokens verified on every agent call, so rules and audit name the person each agent acted for.
Runtime
LangGraph
LangGraph agents with their model calls on the Model Broker and their MCP tools on the Tool Gateway.
Client
Claude Code
Claude Code on a Dome Gateway with per-developer sign-in, rules on every tool call, and audit by name.
Agent service
TinyFish
TinyFish's web agents behind the Tool Gateway, with rules that decide each run on the site it targets.
Talk with ourFDE team.
Or start free, and register your first agent in minutes. Connect it through a gateway, give it rules, and see every call it makes in one audit trail.
Talk with our FDE team
Our engineers design, build and run governed agents for the part of your business you want to change. You see what they deliver.
LLM providers for AI agents: one governed path to every model
Connect a provider once. Its key stays in Dome, and every agent call to it is authorized, metered and audited.