An AI coding assistant for Visual Studio, on your own keys.
Tempr Chat runs inside Visual Studio 2022 (17.14 or later) and Visual Studio 2026: an agent that reads, edits and runs your solution, chat with your code as context, and inline completion, with the model you choose from 39 providers, billed to your own provider account.
7-day free trial on the Standard and Pro plans. The same plan covers VS Code, JetBrains IDEs and the Tempr CLI.
It knows your solution, your debugger and your tests.
Agent mode, your way
A tool-calling agent that reads, searches and edits your code and runs allow-listed commands. Approve every change, auto-apply and review after, or let it run multi-step on Autopilot.
One place to review
Every file the agent touches lands in the Total Changes panel, with a diff per file and Undo All or Accept All.
Debugger and test aware
The agent inspects the live call stack and locals, evaluates expressions mid-session, and runs your tests with structured pass and fail results.
Solution aware
Live compiler diagnostics, project references and the call graph, plus #file, #symbol, @workspace, @terminal and @problems to attach exactly the context you mean.
Semantic code search
Embedding-based search over your solution finds related code even when the wording doesn't match.
Inline completion
Ghost-text suggestions as you type; press Tab to accept.
Personas and delegation
@debugger, @test, @security, @review, @refactor and more, each with its own prompt and tools, and sub-tasks handed to them mid-turn.
MCP servers
Browse a curated catalog of Model Context Protocol servers or add your own, with their tools approval-gated like the built-in ones.
Follows you
Sessions, settings and custom personas sync across machines when you're signed in.
Any model, on your provider account.
39 providers
OpenAI, Anthropic, Google Gemini, xAI, Mistral, DeepSeek, AWS Bedrock, Azure OpenAI and more, or OpenRouter for one key to hundreds of models.
Encrypted, never pooled
Keys are encrypted at rest and scoped to your license. Requests are relayed through Tempr to your provider and billed to your own account, with no markup on tokens.
Your own models
Point Tempr at your own OpenAI-compatible server, such as vLLM, SGLang or LiteLLM, on the Pro and Teams plans.
About Tempr in Visual Studio.
Which versions of Visual Studio does Tempr Chat support?
Visual Studio 2022 version 17.14 or later, and Visual Studio 2026: Community, Professional and Enterprise, on 64-bit Windows.
Which models can I use in Visual Studio?
Any model your provider keys can reach: OpenAI, Anthropic, Google, xAI, DeepSeek, Mistral and the rest of 39 providers, or OpenRouter. On Pro and Teams you can add your own OpenAI-compatible endpoints.
How is Tempr Chat billed?
The Agents plan covers Tempr Chat and the CLI: Standard at $15 a month, Pro at $20, and Teams from $35 a seat. Your provider bills the tokens at its own rates; Tempr adds no markup.
Does it work in VS Code and JetBrains IDEs too?
Yes. The same plan covers Tempr Chat in Visual Studio, VS Code and JetBrains IDEs, and the Tempr CLI in your terminal.
Try it in Visual Studio this week.
Install from the Marketplace, add a key, and ask about your solution.