On this page
Dranz: one route
Dranz keeps threads, permissions, subagents, diffs, review and cost records in one workbench, and there is one place inference goes: Venice's private tier. No provider ladder, no second account. Prompt and response content is not retained after the request under Venice's private-tier contract, and that contract is the whole privacy story for Dranz.
It does not remove operational metadata. Venice can still hold account, billing, model, timestamp, token and IP information.
Set Venice up in Settings › Providers & privacy › Venice. The model picker shows the provider, tier, context window, capabilities and available cost information for every row, so you choose with the route in view.
Reading the badge
Each session carries a badge naming the route that actually ran. Three words do the work:
- Private: Venice's private tier, the route every Dranz session takes. Read it as a provider contract, not an encryption or anonymity claim.
- Anonymized: a separate Venice tier that protects identity, not prompt secrecy. It carries no retention contract.
- E2EE + TEE: available on some Venice models for tool-free turns. A turn that calls structured tools runs on Venice private instead, because the encrypted route does not return tool calls. The badge reports which of the two ran.
Local models (Dranz Pro)
Everything from here down is Dranz Pro. Each Pro route names its provider and privacy tier the same way Dranz names Venice.
Enable Ollama or LM Studio, then open the local models gallery to find models that support tool use and fit the machine's memory. Local is the strongest privacy boundary because inference ends on a service you run. Capability and hardware still matter, so pick a model that fits both the task and the machine.
Direct providers (Dranz Pro)
Connect a Claude subscription, a GLM or Kimi coding plan, or an OpenAI-compatible provider when that route is the right fit. The provider receives the session under its own account and data policy. Dranz names the route without extending Venice's guarantees to it.
Provider choice is fixed when a session starts. If the provider cannot start, Dranz stops and explains the next step instead of falling through to a different service.
Custom profiles (Dranz Pro)
Custom profiles are for providers Dranz does not configure for you, or a subscription that needs its own endpoint and model mapping. Open Settings › Providers & privacy › Custom profiles and add:
- A clear name. Use the provider and plan, such as "GLM Coding Plan" or "Kimi direct", so the route is unmistakable in the session picker.
- The compatible protocol and base URL. Copy the endpoint from the provider's current documentation. Coding-plan endpoints can differ from metered API endpoints, even when both accept the same key.
- The exact model identifier. Model names are provider-defined and case-sensitive.
- A dedicated credential. Keep keys in Dranz's credential flow. Never put a key in a profile name, prompt, project file, issue or support message.
Use Test connection before saving. Dranz validates the route you entered; it never silently replaces a failed custom provider with Venice or another service.
Existing GLM and Kimi plans (Dranz Pro)
GLM Coding Plan and Kimi platform accounts are good custom-profile candidates when you already pay for them or need a specific model. They are direct routes: z.ai or Moonshot receives the session under its own account and data policy. Venice's private-tier terms do not carry across.
Use the endpoint assigned to your plan, follow the provider's current setup guide, then start a short test session and confirm the badge, model and tier before sharing project context.
Hybrid: hosted planning, local execution (Dranz Pro)
A hosted model plans, orchestrates and reviews while a local model handles delegated work on this machine. The hosted part follows its provider's policy; the local part stays local. Dranz labels hybrid sessions and reports when the local route is unavailable.
Before sensitive work
- Confirm the provider and tier in the model picker.
- Read Private as a provider contract, not an encryption or anonymity claim.
- Read Anonymized as identity protection, not prompt secrecy.
- Approved web, search, MCP and shell tools can reach the network independently of model inference.
- Keep provider credentials out of prompts, issues and support conversations.
Asking for Pro
Dranz Pro is in a private build with no public download. Email hello@dranz.ai with what you would do with it. The user study group gets Pro.