Hi NVIDIA team,
Every POST /v1/chat/completions call to moonshotai/kimi-k2.6 returns a 404 “Function not found for account”, even though the key is valid
and the model appears in the catalog. This matches the known “Public API Endpoints not enabled” pattern reported in several recent threads.
Account details
- Account ID: 2q9o6JD5hfCbJfMRz0Ixxl7bPF5cfF5QnOlB9AJqCUY
- API key prefix: nvapi-y_bCq-…
- Endpoint: https://integrate.api.nvidia.com/v1
- Model: moonshotai/kimi-k2.6
- Use case: personal development
What works vs. fails
- GET /v1/models → 200, lists moonshotai/kimi-k2.6 correctly.
- Other models on the same key work fine via POST /v1/chat/completions (e.g. meta/llama-3.1-8b-instruct, nvidia/nemotron-nano-12b-v2-vl,
meta/llama-3.2-90b-vision-instruct). - Only moonshotai/kimi-k2.6 fails.
Reproduce
curl -i --max-time 15 “https://integrate.api.nvidia.com/v1/chat/completions”
-H “Authorization: Bearer $NVIDIA_API_KEY”
-H “Content-Type: application/json”
-d ‘{“model”:“moonshotai/kimi-k2.6”,“messages”:[{“role”:“user”,“content”:“hi”}]}’
Response
{“status”:404,“title”:“Not Found”,“detail”:“Function ‘23d4f03a-b8a6-4adb-a183-7daa083a09cc’: Not found for account
‘2q9o6JD5hfCbJfMRz0Ixxl7bPF5cfF5QnOlB9AJqCUY’”}
The same function UUID (23d4f03a-…) 404s across multiple accounts in other threads, so this looks like the kimi-k2.6 endpoint isn’t mapped
for my account. Could you please enable Public API Endpoints / provision this model’s function for my account? Thanks!