New
- Tenant Workspace Templates. MSP admins can now define a default configuration for every new tenant — including which AI model they start on, which models are available or blocked, whether memory, personalization, and subagents are enabled, and which apps and agents are pre-loaded. Create multiple templates for different customer types and choose the right one at provisioning time. Set it once, and every new tenant starts from the same baseline.
- Share apps, agents, and workflows with all current and future tenants in one toggle. From the Share Settings panel on any Workshop item, turn on Share with all current and future tenants and you're done. New tenants are automatically granted access the moment they're provisioned — no manual assignment required.
- Control who can edit, not just who can access. Enabling Allow internal organization to edit gives Workshop creation rights to internal team members — not tenant users — to make changes to a shared item. Opt-in per item, so you stay in control of what's editable and what's locked down.
- LLM Gateway expanded to include Anthropic messages compatibility. If you're already using the Anthropic SDK or building with Claude, you can now send requests in the Anthropic messages format directly to Hatz. Your calls route through Hatz's model system — usage tracked, billed, and governed the same way as everything else. It's not a full Anthropic API passthrough; it covers the messages endpoint.
- New LLM Opus 4.8 added to the model selector for chat, apps, agents, and workflows.
One of the most prominent improvements in Opus 4.8 is its honesty. We train all our models to be honest—for instance, to avoid making claims that they can’t support. But a general problem with AI models is that they sometimes jump to conclusions, confidently claiming to have made progress in their work despite the evidence being thin. Early testers report that Opus 4.8 is more likely to flag uncertainties about its work and less likely to make unsupported claims. This is borne out in our evaluations, which show that Opus 4.8 is around four times less likely than its predecessor to allow flaws in code it has written to pass unremarked.
- Anthropic
Improved
- Improved handling of oversized documents for greater reliability.
Fixed
- Fixed an issue impacting select users where images attached to a conversation weren't being read correctly.