Inspect configured Kimi Code providers
Start with the provider list before adding or changing credentials.
kimi provider list
kimi provider catalog list --filter kimi
kimi provider catalog list --filter openai
API key boundary
Understand whether Kilo Code API keys work in Kimi Code, how provider keys differ, and what to check before configuring third-party tools.
| Check | Pass condition | If it fails |
|---|---|---|
| Issuer | The key issuer supports the endpoint you configure | Get a key from the correct provider |
| Base URL | The endpoint matches the tool protocol | Do not reuse the key |
| Model ID | The exact model ID is supported | Switch to a listed model |
| Billing | Usage is charged to the intended account | Stop before running long tasks |
Do not assume a Kilo Code key works in Kimi Code. API keys belong to a provider, endpoint, and permission model. Kimi Code has its own documented provider and model configuration rules, and official Kimi Code API setup uses Kimi Code Console, base URLs, and exact model IDs.
If Kilo Code is acting as a provider or aggregator, check whether it exposes an OpenAI-compatible or Anthropic-compatible endpoint that Kimi Code can be configured to call. If it does not, the key is not interchangeable.
Do not paste a key into Kimi Code until the issuer, endpoint, protocol, and model ID match.
Start with the provider list before adding or changing credentials.
kimi provider list
kimi provider catalog list --filter kimi
kimi provider catalog list --filter openai
Use this only when the key issuer documents an OpenAI-compatible endpoint.
export KIMI_MODEL_NAME="<model-id>"
export KIMI_MODEL_API_KEY="<provider-api-key>"
export KIMI_MODEL_BASE_URL="<provider-base-url>"
export KIMI_MODEL_PROVIDER_TYPE="openai"
kimi -p "Reply only: PROVIDER_READY"
This catches a bad base URL or key before a long coding session.
BASE_URL="<provider-base-url>"
PROVIDER_API_KEY="<provider-api-key>"
curl -s "$BASE_URL/models" -H "Authorization: Bearer $PROVIDER_API_KEY"
Use this when you are done with the temporary provider test.
unset KIMI_MODEL_NAME KIMI_MODEL_API_KEY KIMI_MODEL_BASE_URL KIMI_MODEL_PROVIDER_TYPE
Use this section to keep the next step practical before you touch accounts, files, infrastructure, or hardware.
| Check | Look for | Why it matters |
|---|---|---|
| Start narrow | Run the smallest read-only check before commands that install, move, write, or call a provider. | A small pass condition keeps the workflow reviewable. |
| Keep commands copyable | Use exact commands without shell prompts and run one step at a time. | This avoids accidental pasted characters or bundled side effects. |
| Protect accounts | Do not paste raw API keys, billing data, OAuth codes, or private files into a guide page. | Account-bound work belongs in the official product surface. |
| Record the result | Write down the model ID, path, version, output, or error message that changed the decision. | The next attempt should not have to rediscover the same state. |
Find who issued the key, what endpoint it authorizes, and which models it can call.
Kimi Code third-party setup documents OpenAI-compatible and Anthropic-compatible base URLs for Kimi Code API.
A key that works in one tool can fail or leak if pasted into an unrelated provider field.
After configuration, run a tiny model check before using the key in a long coding session.
The same model family can appear behind different endpoints, auth headers, billing accounts, and rate limits.
Use documented IDs such as k3, k3-256k, kimi-for-coding, or kimi-for-coding-highspeed when configuring Kimi Code.
Kimi Code documents how configuration and credentials are stored. Follow that path rather than hardcoding keys in prompts or shell history.
Only if the key's provider, endpoint, protocol, and model IDs are supported by the Kimi Code configuration you are using.
Run a small direct model request after configuration and confirm it bills the intended account.
Use the documented provider and model configuration path, not a chat prompt.