Inference included
Use powerful AI models without buying GPUs, negotiating separate provider contracts, or operating model infrastructure.
CostIQ Inference API
Powerful AI inference for your whole company.
CostIQ provides the compute and a simple API. We provision separate keys for your company, teams, employees, and applications—each with its own model access, usage limits, and reporting.
Your companyOrganization API access
Engineering
ciq_••••7f2acodeCustomer support
ciq_••••91c4general-fastProduction app
ciq_••••2be7general-fastSeparate keys · Independent limits · One usage report
What you get
One AI inference service for your entire organization.
Separate API keys
Give each company, team, employee, application, and agent its own revocable CostIQ credential.
Usage controls
Set model access, expiry, request rates, token rates, concurrency, and estimated-spend limits for every key.
Clear reporting
See requests, tokens, latency, and estimated cost by customer, team, employee, or workload.
How it works
From request to inference in three steps.
Tell us what you need
Share your use case, expected usage, preferred models, and who needs access.
Receive your organization and keys
We provision CostIQ access and create separate credentials with the right limits for each approved user or application.
Start using inference
Connect coding agents, tools, and products through native Responses or Chat Completions APIs, then monitor usage as your organization grows.
Use the API
One endpoint for the work your company already does.
Use a CostIQ key with Codex, standard HTTP requests, and compatible clients. The same API can serve people, internal tools, agents, and products.
Coding agentsCustomer supportInternal toolsDocument analysisResearchAutomated workflowsBatch processing
POST /v1/responses
export COSTIQ_BASE_URL="https://inference.costiq.xyz/v1"
curl "$COSTIQ_BASE_URL/responses" \
-H "Authorization: Bearer $COSTIQ_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"model": "general-fast",
"input": "Help me understand this repo",
"store": false,
"stream": true
}'Get started
Give your company access to AI inference.
Tell us how many people or applications need access and what you want to run.