Coding Agents

Roo Code setup with Thalam

Install Roo Code, paste your Thalam API key into the OpenAI-Compatible provider, and ship in under a minute. Claude, GPT, Gemini, DeepSeek, Qwen, Kimi and more — all in the same VS Code sidebar, all from one API key.

Available inVS Code
Install Roo Code

Setup in three steps

Roo Code uses the same provider picker as Cline — pick "OpenAI Compatible" and fill in three fields.

1

Sign up

Create a THALAM account and copy your API key from the dashboard. No credit card required.

2

Open Roo Code settings

In your editor open the Roo Code sidebar and click the gear icon at the top of the panel.

3

Choose "OpenAI Compatible"

Pick "OpenAI Compatible" from the API Provider dropdown, then fill in the three fields below.

Roo Code → Settings

API Provider

OpenAI Compatible

Pick this from the dropdown.

Base URL

https://api.thalam.ai/v1

Copy verbatim — must include /v1.

API Key

tl-live-...

Paste the key from your THALAM dashboard.

Model ID

anthropic/claude-sonnet-4.6

Any model id from the THALAM catalog. The model must support tool calling for Roo Code's agent loop to work — flagship Claude / GPT / Gemini / DeepSeek-V3 / Qwen3 all do.

Browse every available model id on the (format is <provider>/<model>).

Verify it works

Open a new Roo Code task and ask:

Which model are you and which API base URL is serving you?

Common gotchas

Three issues that account for almost every "it didn't work" report.

Model can't call tools

Per Roo Code's docs, the agent loop requires OpenAI-compatible tool calling. If you pick a smaller open-weight model that doesn't expose tools, Roo will refuse it. Use a flagship (Claude / GPT / Gemini) or a tool-calling-capable open-weight model (DeepSeek V3.2, Qwen3 series).

404 model not found

The Model ID field expects the full id <provider>/<model> — for example anthropic/claude-sonnet-4.6, not claude-sonnet-4.6 alone.

Connection error

The Base URL must include the /v1 suffix and use https. Paste it as-is: https://api.thalam.ai/v1.


Why use THALAM with Roo Code

Roo Code is built around an autonomous agent loop — model choice and per-call cost matter more than in any other tool.

Every leading model, one key

Claude, GPT, Gemini, DeepSeek, Qwen, Kimi, Llama, Mistral, GLM. Switch the Model ID field and the next agent step routes to a different upstream — no extra setup.

Routine work on cheap models

Use DeepSeek V3.2 or Qwen3 for repetitive multi-step refactors, and reserve Claude or GPT flagships for the steps that actually need deep reasoning.

No lock-in, pay per token

No subscription, no minimums. To swap THALAM for a provider-direct setup, change Base URL and API Key — Roo Code's UX stays identical.

OpenAI-compatible end to end

Streaming, tool calling, structured outputs, vision — every Roo Code feature works through Thalam's /v1 endpoint without custom adapters.


Frequently asked questions

What developers ask before wiring Thalam into Roo Code.

Do I need an API key to use Roo Code with Thalam?

?

Yes. Roo Code calls models through whichever provider you select in its settings. To route Roo Code through Thalam, create a free Thalam account, copy your API key from the dashboard, and paste it into Roo Code's API Key field after picking "OpenAI Compatible". No credit card required to create the key — you only pay when you make requests.

How do I install Roo Code and connect it to Thalam?

?

Three steps. (1) Install the Roo Code extension from the VS Code Marketplace. (2) Sign up at thalam.ai and copy your tl-live-... key. (3) Open the Roo Code sidebar, click the gear icon, pick "OpenAI Compatible" from the API Provider dropdown, set Base URL to https://api.thalam.ai/v1, paste your key, and choose a tool-calling-capable model id like anthropic/claude-sonnet-4.6.

Is Thalam OpenAI-compatible?

?

Yes. Thalam exposes an OpenAI-compatible base URL at https://api.thalam.ai/v1, so Roo Code's "OpenAI Compatible" provider treats it the same way it would treat OpenAI direct — just with a different Base URL and API Key. Every model Thalam routes to (Claude, Gemini, DeepSeek, Qwen and more) is reachable through the same OpenAI Chat Completions interface.

Can I use Thalam with Roo Code on Mac and Windows?

?

Both, plus Linux. Roo Code runs as a VS Code extension, and VS Code is available on macOS, Windows and Linux. The Thalam setup (Base URL + API Key + Model ID) is identical on every platform.

Can I switch models without changing my code?

?

Yes. Change the Model ID field in Roo Code's settings to any other Thalam model id (provided it supports tool calling — Roo's agent loop requires it) and the next task routes to that model. Switching between Claude, GPT, DeepSeek-V3 or any other tool-calling Thalam-routed model is a config change, not a code change.

Plug in. Pick a model. Ship.

Get an API key in under a minute. No credit card required.