Last updated September 14, 2026

Venice AI in Codex — Private Models | Codex Router

Venice in Codex means adding Venice AI’s private models to the Codex model picker through Codex Router. Provider id venice points at https://api.venice.ai/api/v1, but a key alone is not enough: Venice API access is an entitlement.

What the Venice route is

The venice provider is an API-key provider with pinned routes on top of a curatable catalog. A key from venice.ai/settings/api authenticates the endpoint. The router prints the entitlement requirement when you connect the provider rather than letting it arrive as a 403 inside Codex.

Access is an entitlement

A free Venice account has no API access. One of the following is what makes the key usable:

Access path	What it means
Pro subscription	The low-rate-limit Explorer tier
Funded USD balance	Prepaid API credit
Staked VVV	Staking that grants VCU

Without one of these, requests fail authorization. The router surfaces that requirement at connect time so the failure does not appear mid-task.

Enable Venice

Install Codex Router first (see the install guide), then store the key, enable the provider, and curate:

./bin/model-router codex provider-key venice set
./bin/model-router codex providers enable venice
./bin/curate-models venice

On Windows use ./model-router.ps1 codex with the same commands. Then fully quit Codex, reopen it, and create a new task.

Pinned routes and curation

The pinned routes include Claude Fable 5.1 and, historically, GLM-5.3-Flash. The Ox Alpha preset was withdrawn after wire verification was billing-blocked, so it is not shipped. Additional models come from the live catalog through curation.

Route	Status
Claude Fable 5.1	Pinned
GLM-5.3-Flash	Historically pinned
Ox Alpha preset	Withdrawn — verification was billing-blocked

Curation stores what you select in protected state, and each entry keeps the metadata you provide or the catalog advertises.

Verify the integration

./bin/model-router codex doctor
./bin/model-router codex providers
./bin/test-model 'venice/MODEL_ID' --live --yes

The provider must report SHOW and ready. If entries are missing, see models missing from the picker.

Provider policy

Venice may restrict or block proxy traffic, and access remains an entitlement controlled by Venice. Routing its models through Codex Router is at your own risk; check Venice’s terms before relying on it.

FAQ

What is the Venice route in Codex? It is Codex Router’s venice provider, which routes Venice AI models into the Codex picker from https://api.venice.ai/api/v1 using a Venice API key.

Does a free Venice account work? No. Venice API access is an entitlement, not just a key. A Pro subscription, a funded USD balance, or staked VVV that grants VCU is what makes the key usable.

Can I add more Venice models? Yes. After enabling the provider, run ./bin/curate-models venice to select additional models from the live catalog.