Use CoralBricks with GitHub Copilot
Bring CoralBricks open models — GLM and Kimi with up to 1M context — into GitHub Copilot Chat in VS Code with Bring Your Own Key.
GitHub Copilot's Bring Your Own Key (BYOK) lets you point Copilot Chat at
any OpenAI-compatible endpoint. CoralBricks plugs in as a custom provider —
Copilot's chat, edits, and agent mode then run on GLM 5.2 or Kimi K3, billed
to your CoralBricks account instead of your Copilot quota.
Requires a paid Copilot plan (Business or Enterprise) and VS Code. BYOK
is a Copilot platform feature; on plans without it, use
OpenCode or Cline instead. BYOK also does
not apply to inline code completions — those stay on Copilot's own models
(a Copilot platform restriction, same as Cursor's Tab).
1. Get an API key
Create a key from your API keys page (format cb_...).
2. Add CoralBricks as a custom model
In VS Code:
- Open the Chat view and run Chat: Manage Language Models from the
Command Palette (or click the model picker → Manage Models).
- Choose the OpenAI Compatible provider.
- Fill in:
- Base URL →
https://inference.coralbricks.ai/v1
- API Key → your
cb_... key
- Model ID →
glm-5.2-fp4 (or any slug from the table below)
- Confirm. The model appears in the Copilot Chat model picker.
Add more than one model by repeating step 2–3 with a different slug.
Available models
| Model |
Model ID |
Context |
Input $/M |
Output $/M |
| GLM 5.2 |
glm-5.2-fp4 |
1M |
$1.12 |
$4.40 |
| Kimi K3 |
kimi-k3 |
1M |
$3.00 |
$15.00 |
| GPT-OSS 120B |
gpt-oss-120b |
128K |
$0.12 |
$0.60 |
Cached input tokens are always free.
What routes through CoralBricks
Copilot Chat, Edits, and Agent mode for the model you select in
the picker — including tool calls in agent mode. Inline completions (the
grey ghost text) always use Copilot's own backend; BYOK does not cover them.
Usage is billed by CoralBricks and does not count against your Copilot
request quota.
Notes
- The API is OpenAI wire-format at
https://inference.coralbricks.ai/v1 — the
same key also works in OpenCode, Cline,
Cursor, and anything else that speaks OpenAI-compatible
providers. See the API reference for the full surface.
- Long context is the point: GLM 5.2 and Kimi K3 accept up to 1M tokens, so
large-repo Copilot Chat context fits without truncation.