We've been busy since our release of LLM Gateway 2.0 two weeks ago, and we're excited to announce some new improvements:

New Model Support

We've added support for several new models from multiple providers:

• Claude Haiku 5.5, Sonnet 5.5, Fable 5 and Fable 5.1

• GPT 6.1 Sol

• Grok 4.7

• GLM 5.3

• Nova 2 multimodal embeddings

To see a full list of current supported models, check our docs: Supported Models.

Per-key Spend Limits

Now, you can set monthly spend limits on individual keys, in addition to your global spend limit. A per-key spending limit caps what a single key can spend each month. It keeps one application, agent, or teammate from using up your organization's whole budget. We've also added visual indicators to the LLM Gateway page so you can see how close each key is to reaching its spend limit. To learn more, see Cost Visibility & Control.