Available Models

List Models Available to Your Key

curl "https://api.llmnex.com/v1/models" \
  -H "Authorization: Bearer YOUR_API_KEY"

The default endpoint is https://api.llmnex.com. For Hong Kong acceleration, use https://hk.llmnex.com/v1/models with the same authentication. See API endpoints.

Only Gemini models are currently supported. Use the full model names returned by the list; permissions may differ between keys.

The new gemini-nano-banana-2.1 model has no -c suffix and uses token-based billing. Check that your key has access before using it, and consult the model-specific documentation for capabilities.

Model Names & Billing

Model name	Billing	Example
With -c suffix	Per-request	gemini-3-pro-image-c
Without -c suffix	Token-based	gemini-3-pro-image

The suffix distinguishes billing modes, not fixed resolutions. The old fixed-resolution-suffix models have been retired; do not use them.

These names illustrate the naming convention. Actual availability depends on the model list for your key. Prices and actual charges are determined by the console and your account bill. usageMetadata reports usage and cannot by itself identify the billing mode.

For resolution parameters, see Models & Resolution.

When a Model Is Unavailable

For model_not_found or No available channel for model ..., check spelling and whether the model has been retired, then query the models available to your key. Names with and without -c identify different models; do not swap them automatically without confirming the billing mode.

If the required model is missing from the list, contact the service provider to confirm access. If it is listed but still fails, include X-Oneapi-Request-Id in your report.