DeepInfra API and Dome
Open models on DeepInfra, with every call governed. DeepInfra hosts open models like DeepSeek and Gemma under one key. Connect it to Dome, tag connections with the region you approved, and an agent is refused on anything untagged. Together AI serves the same DeepSeek V4 Flash as a fallback.
Dome provides model brokeringand routing for DeepInfra.
The key stays put
Dome keeps DeepInfra's key and adds it to each outbound request. Agents hold Dome keys only, so a leaked agent key never exposes DeepInfra's.
Two hosts, one release
Together AI serves the same DeepSeek V4 Flash release. Put both in a pool and DeepInfra's failures land on Together.
Your tags, your rules
Tag each connection with the region you approved, and a rule refuses anything without the tag.
Connect DeepInfrain three steps.
Add DeepSeek V4 Flash on DeepInfra, pair it with Together AI, then hand agents the Gateway address.
Add the connection
The provider id is deepinfra. Dome uses https://api.deepinfra.com/v1/openai and DeepInfra's own model ids.
shelldome models add deepseek-deepinfra \ --provider deepinfra \ --model deepseek-ai/DeepSeek-V4-Flash-0731 \ --api-key "$DEEPINFRA_API_KEY" \ --attributes '{"region":"us"}' \ --gateway prod-gatewayPool it with Together AI
Add the same DeepSeek id on Together as a second connection, with DeepInfra going first.
shelldome models pool create deepseek-v4-flash \ --failover-max all --gateway prod-gateway dome models pool member add deepseek-v4-flash deepseek-deepinfra --priority 0dome models pool member add deepseek-v4-flash deepseek-together --priority 1Point your agent at the Gateway
Keep your OpenAI-compatible code. Point it at the Gateway and hand it the agent's Dome key.
pythonfrom openai import OpenAI client = OpenAI(base_url=f"{GATEWAY_URL}/v1", api_key=DOME_AGENT_KEY)client.chat.completions.create( model="deepseek-v4-flash", messages=[{"role": "user", "content": "Summarize this claim file."}],)
Approvedregions only.
Applied to an agent, this refuses any connection not tagged region: us. A new connection stays out of reach until someone tags it.
cedar
forbid (principal, action == Dome::Action::"llm:invoke", resource is Dome::LLMModel)unless { resource has region && resource.region == "us" };Bringing it alltogether.
Connecting DeepInfra 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
Common questionsabout DeepInfra.
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.