Skip to content

fix: smart vision setup that respects the user's chosen provider#945

Closed
teknium1 wants to merge 1 commit into
mainfrom
fix/smart-vision-setup
Closed

fix: smart vision setup that respects the user's chosen provider#945
teknium1 wants to merge 1 commit into
mainfrom
fix/smart-vision-setup

Conversation

@teknium1

Copy link
Copy Markdown
Contributor

Problem

The setup wizard blindly asks for an OpenRouter API key after ANY non-OpenRouter provider selection — even for Nous Portal and Codex which already support vision natively. If you just chose OpenAI as your provider, getting Enter OpenRouter API key thrown in your face is confusing and annoying.

Solution

Replace the blanket OpenRouter key prompt with provider-aware vision setup:

Provider chosen Vision? What happens
OpenRouter Skip — Gemini via OR key
Nous Portal OAuth Skip — Gemini via Nous
OpenAI Codex Skip — gpt-5.3-codex handles vision
Custom (api.openai.com) Show OpenAI vision model picker (gpt-4o, gpt-4o-mini, gpt-4.1, etc.)
Custom (other) Check existing creds, then offer: OpenRouter / OpenAI / Skip
z.ai / kimi / minimax / nous-api Same — check existing creds first, then friendly choice

For providers that don't support vision, the prompt is friendlier and contextual:

◆ Vision & Image Analysis (optional)
  Vision requires a multimodal-capable provider. Z.AI / GLM
  doesn't natively support it. Choose how to enable vision,
  or skip to configure later.

→ OpenRouter — uses Gemini (free tier at openrouter.ai/keys)
  OpenAI — enter API key & choose a vision model
  Skip for now

Also fixes the setup summary to check actual vision availability across all providers instead of hardcoding 'requires OPENROUTER_API_KEY'. MoA still correctly requires OpenRouter (calls multiple frontier models).

Changes

  • hermes_cli/setup.py: Replaced vision setup block + fixed summary section

The old flow blindly asked for an OpenRouter API key after ANY non-OR
provider selection, even for Nous Portal and Codex which already
support vision natively. This was confusing and annoying.

New behavior:
- OpenRouter: skip — vision uses Gemini via their OR key
- Nous Portal OAuth: skip — vision uses Gemini via Nous
- OpenAI Codex: skip — gpt-5.3-codex supports vision
- Custom endpoint (api.openai.com): show OpenAI vision model picker
  (gpt-4o, gpt-4o-mini, gpt-4.1, etc.), saves AUXILIARY_VISION_MODEL
- Custom (other) / z.ai / kimi / minimax / nous-api:
  - First checks if existing OR/Nous creds already cover vision
  - If not, offers friendly choice: OpenRouter / OpenAI / Skip
  - No more 'enter OpenRouter key' thrown in your face

Also fixes the setup summary to check actual vision availability
across all providers instead of hardcoding 'requires OPENROUTER_API_KEY'.
MoA still correctly requires OpenRouter (calls multiple frontier models).
@teknium1

Copy link
Copy Markdown
Contributor Author

Merged via #1323. I cherry-picked the substantive setup wizard change onto current main, then added a small follow-up so the default OpenAI vision choice persists AUXILIARY_VISION_MODEL and added regression coverage for the new flow.

@teknium1 teknium1 closed this Mar 14, 2026
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant