[NemoClaw][MacOS][Brev] [tag v0.0.1] Base URL example is misleading as per other OpenAI-compatible endpoint

[caroline-xuan]

Description

[Description]
[tag v0.0.1] nemoclaw onboard wizard, already specified correct openrouter api key

[3/7] Configuring inference (NIM)
──────────────────────────────────────────────────

Inference options:
1) NVIDIA Endpoints (recommended)
2) OpenAI
3) Other OpenAI-compatible endpoint
4) Anthropic
5) Other Anthropic-compatible endpoint
6) Google Gemini
7) Local Ollama (localhost:11434)

Choose [1]: 3
OpenAI-compatible base URL (e.g., https://openrouter.ai/api/v1): https://openrouter.ai/api/v1
Other OpenAI-compatible endpoint model []: qwen/qwen3-next-80b-a3b-instruct:free
Other OpenAI-compatible endpoint endpoint validation failed.
Responses API: HTTP 429: Provider returned error | Chat Completions API: HTTP 429: Provider returned error
Please enter a different Other OpenAI-compatible endpoint model name.

[Environment]

Item	Version / detail
Device	macos
Node.js	v22.x
OpenShell CLI	0.0.16
NemoClaw	v0.1.0
OpenClaw	2026.3.11 (29dc654)

[Steps to Reproduce]

1. nemoclaw onboard, select 3) Other OpenAI-compatible endpoint for inference
2. When prompting: OpenAI-compatible base URL (e.g., https://openrouter.ai/api/v1): fill https://openrouter.ai/api/v1 as shown and a correct api key
3. When prompting Other OpenAI-compatible endpoint model []: fill qwen/qwen3-next-80b-a3b-instruct:free


[Expected Result]
The base URL hint should be accurate, i.e. only the https+hostname without /api/v1

[Actual Result]

Choose [1]: 3
OpenAI-compatible base URL (e.g., https://openrouter.ai/api/v1): https://openrouter.ai/api/v1
Other OpenAI-compatible endpoint model []: qwen/qwen3-next-80b-a3b-instruct:free
Other OpenAI-compatible endpoint endpoint validation failed.
Responses API: HTTP 429: Provider returned error | Chat Completions API: HTTP 429: Provider returned error
Please enter a different Other OpenAI-compatible endpoint model name.

Other OpenAI-compatible endpoint model []:

# start over and fill the URL without /api/v1
nemoclaw onboard

NemoClaw Onboarding

[1/7] Preflight checks
──────────────────────────────────────────────────
✓ Docker is running
✓ Container runtime: colima
✓ openshell CLI: openshell 0.0.16
✓ Port 8080 already owned by healthy NemoClaw runtime (OpenShell gateway)
✓ Port 18789 already owned by healthy NemoClaw runtime (NemoClaw dashboard)
✓ Apple GPU detected: Apple M4 Pro (20 cores), 49152 MB unified memory
ⓘ NIM requires NVIDIA GPU — will use cloud inference
Reusing healthy NemoClaw gateway.

[3/7] Configuring inference (NIM)
──────────────────────────────────────────────────

Inference options:
1) NVIDIA Endpoints (recommended)
2) OpenAI
3) Other OpenAI-compatible endpoint
4) Anthropic
5) Other Anthropic-compatible endpoint
6) Google Gemini
7) Local Ollama (localhost:11434)

Choose [1]: 3
OpenAI-compatible base URL (e.g., https://openrouter.ai/api/v1): https://openrouter.ai
Other OpenAI-compatible endpoint model []: qwen/qwen3-next-80b-a3b-instruct:free
Responses API available — OpenClaw will use openai-responses.
Using Other OpenAI-compatible endpoint with model: qwen/qwen3-next-80b-a3b-instruct:free

[4/7] Setting up inference provider
──────────────────────────────────────────────────
✓ Active gateway set to 'nemoclaw'
✓ Created provider compatible-endpoint

Steps to reproduce

No steps provided.

---

[NVB# 6035277]

[NVB#6035277]

[wscurran]

✨ Thanks for reporting this documentation issue, which identifies a problem with the base URL example being misleading, and could help us improve the clarity and accuracy of our documentation and onboarding process.

---

Possibly related open PRs:

• #1226 docs: use CUDA base image for GPU smoke test in spark-install
• #1167 fix: show macOS launchctl instructions instead of systemctl on darwin (#1067)
• #1235 refactor(cli): delete setup.sh, route all setup through nemoclaw onboard