Skip to content
NexInferNexInfer Docs

Codex

Configure

Section titled “Configure”
Terminal window
export OPENAI_BASE_URL=https://api.nexinfer.com/v1
export OPENAI_API_KEY="$NEXINFER_API_KEY"
codex

Use a key permitted for GPT/Codex models. If your Codex version uses a provider block, set its base URL to the same value and select the Responses API.

Verify

Section titled “Verify”
Terminal window
curl https://api.nexinfer.com/v1/responses \
  -H "Authorization: Bearer $NEXINFER_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"YOUR_MODEL_ID","input":"Reply with OK."}'

Then run a Codex task that requires a tool call. NexInfer supports both /v1/responses and the legacy /responses compatibility path.

NexInfer does not intentionally change tools, reasoning fields, service tier, session fields, context, or the requested model on the dedicated passthrough channel.