Coding Agents

Cursor setup with Thalam

Install Cursor, paste your Thalam API key, and ship in under a minute. Override the OpenAI Base URL to https://api.thalam.ai/v1 and Claude, GPT, Gemini, DeepSeek, Qwen and more appear in Cursor's model picker.

Get Cursor

Read this first — what BYOK covers in Cursor

Per Cursor's official docs: "Custom API keys only work with chat models. Tab completion continues using Cursor's built-in models."

In practice this means routing chat through THALAM works today; Cursor Tab autocomplete stays on Cursor's own infrastructure. If you want every Cursor feature — chat, Tab, Composer, Agent — running through THALAM, use or instead — both work inside Cursor and route everything through your key.


Setup in three steps

Cursor configures custom models through its Settings UI — no config files.

1

Sign up

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

2

Open Cursor settings

Press Cmd+, (Mac) or Ctrl+, (Windows/Linux) and go to the Models tab.

3

Add THALAM as your provider

Paste your THALAM key into the OpenAI API Key field, enable "Override OpenAI Base URL", and add the model ids you want to use.

Cursor Settings → Models

OpenAI API Key

tl-live-...

Paste your THALAM API key here. Cursor calls this field "OpenAI API Key" because THALAM is OpenAI-compatible — that's expected.

Override OpenAI Base URL

https://api.thalam.ai/v1

Enable the toggle and paste the URL exactly (must include /v1). Click "Verify" — Cursor will reach out to the endpoint and confirm the key works.

Custom models

anthropic/claude-sonnet-4.6

In the Models list, add each model id you want available in the picker. Use the format <provider>/<model> — e.g. anthropic/claude-sonnet-4.6, deepseek/deepseek-v3.2, openai/gpt-5.

Browse available model ids on the . You can add as many models as you want — each one appears in Cursor's model picker after saving.

Verify it works

Open a new Cursor chat (Cmd+L / Ctrl+L), pick a THALAM model from the picker, 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.

Verify button fails

Confirm the Base URL ends with /v1 and the key starts with tl-live-. Cursor's Verify call hits /models on the URL you provide; both must be correct or it fails.

Model not in picker

You haven't added the model id to the Models list yet. Cursor doesn't auto-discover — you list each model id you want available. Use <provider>/<model>.

Cursor's built-ins broke

Cursor's BYOK currently sends all "openai-family" traffic to the overridden URL. Built-in Cursor models (Claude, Gemini) keep working through Cursor; only GPT-style models route through THALAM.


Why use THALAM with Cursor

You keep Cursor's UI and Tab autocomplete; chat routes through THALAM so you can pick any model and pay per token.

Models Cursor doesn't ship

Add DeepSeek, Qwen, Kimi, GLM, MiniMax, Llama variants — any model THALAM lists. Use them in Cursor chat just like the built-ins.

Pay per token

No fixed subscription on the chat side. Heavy users pay more, light users pay less — same THALAM dashboard tracks usage across every tool you use.

One key across tools

The same THALAM key works in Cursor, Cline, Continue, OpenClaw, Aider, OpenCode, Zed — wire it once per tool, manage usage in one dashboard.

Multimodal upstream

Cursor chat is text-only today, but the same THALAM key reaches image, video, audio, and embedding models from other tools or your own scripts.


Frequently asked questions

What developers ask before wiring Thalam into Cursor.

Do I need an API key to use Cursor with Thalam?

?

Yes. Cursor calls models through whichever provider you wire into its OpenAI API Key field. To route Cursor chat through Thalam, create a free Thalam account, copy your API key from the dashboard, and paste it into Cursor's OpenAI API Key field with the Base URL override enabled. No credit card required to create the key — you only pay when you make requests.

How do I install Cursor and connect it to Thalam?

?

Three steps. (1) Install Cursor from cursor.com on Mac, Windows or Linux. (2) Sign up at thalam.ai and copy your tl-live-... key. (3) Open Cursor settings (Cmd+, on Mac, Ctrl+, on Windows/Linux), go to the Models tab, paste your key into "OpenAI API Key", enable "Override OpenAI Base URL" and set it to https://api.thalam.ai/v1, then add the model ids you want.

Is Thalam OpenAI-compatible?

?

Yes. Thalam exposes an OpenAI-compatible base URL at https://api.thalam.ai/v1, so Cursor treats it the same way it treats OpenAI — just with a different baseUrl and apiKey. 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 Cursor on Mac and Windows?

?

Both, plus Linux. Cursor runs natively on macOS, Windows and Linux; the Base URL override and API Key live in Cursor's in-app Settings UI on every platform, so the Thalam setup is identical regardless of OS.

Can I switch models without changing my code?

?

Yes. Add as many model ids as you want to Cursor's Custom models list, then pick from Cursor's model picker at runtime. Switching between Claude, GPT, DeepSeek or any other Thalam-routed model is a config change, not a code change — Cursor calls the same /v1 endpoint either way.

Plug in. Pick a model. Ship.

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