Skip to content

Lainnya

Use this page for OpenAI-compatible IDEs, CLIs, and gateways that do not have a dedicated Client Integrations guide. VS Code custom-model setups are out of scope for LootVM v1.

  1. Create a LootVM API key.
  2. Set the client’s OpenAI-compatible base URL to:
https://api.lootvm.com/v1
  1. Set the client’s API key field to the LootVM key (Bearer).
  2. Set the model to an exact id from GET /v1/models for that key.
  3. Send a short non-stream chat completion, then enable streaming if the tool supports it.

Do not append /chat/completions to the base URL unless the tool’s docs say the field is a full path (most clients append the path themselves).

Terminal window
export LOOTVM_KEY='your-lootvm-api-key'
curl --fail-with-body https://api.lootvm.com/v1/models \
-H "Authorization: Bearer ${LOOTVM_KEY}"
curl --fail-with-body https://api.lootvm.com/v1/chat/completions \
-H "Authorization: Bearer ${LOOTVM_KEY}" \
-H "Content-Type: application/json" \
-d '{
"model": "your-enabled-model",
"messages": [{"role": "user", "content": "Hello from LootVM"}]
}'

If curl works and the tool fails, the tool’s base URL, auth header name, or model field is wrong—not the gateway.

If the tool speaks Anthropic Messages instead of Chat Completions, use the Claude CLI pattern or Messages (Claude): host https://api.lootvm.com, path /v1/messages, LootVM key as Bearer or x-api-key.

Symptom Check
401 Key not sent as Bearer, or not a LootVM-issued key
404 Base URL missing /v1, or /v1 doubled with an extra path
Model errors Id mismatch vs /v1/models; group/model not enabled for the key
Tool still bills OpenAI / Anthropic directly Override/base URL setting was not saved or not applied to the active profile

For SDK-level copy-paste, see Foundation (SDKs).