vlm

API-backed vision-language Q&A. Zero GPU: every provider is a remote endpoint. Images are gap-native uint8 [H, W, 3] numpy arrays, PNG-encoded on the wire.

Providers

Selected by GAP_VLM_PROVIDER (default openrouter); each tool also accepts a per-call provider= override.

Provider	Backend	Config (env)
openrouter	OpenRouter's OpenAI-compatible chat-completions API	OPENROUTER_API_KEY (or GAP_VLM_API_KEY); GAP_VLM_MODEL (default gemini-3.1-flash-lite-preview, see DEFAULT_MODEL in tools.py); set GAP_VLM_BASE_URL to point at another OpenAI-compatible server (e.g. a local vLLM)
vertex	Vertex AI via google-genai (Gemini models)	GAP_VLM_MODEL, GAP_VLM_PROJECT_ID, GAP_VLM_REGION

The vertex provider lazy-imports google-genai — install the engine's vertex extra first: pip install "graph-as-policy[vertex]".

When to use

• Semantic scene checks and checkpoint verification (vlm.query_yes_no).
• Free-form scene descriptions or attribute queries (vlm.query).
• Prefer gemini-er.detect when you need pixel-space bounding boxes, and molmo.point_prompt when you need a single click point.

Notes

• vlm.query_yes_no coerces with the source-verbatim rule: answer is true iff "yes" appears in the lowercased reply.
• Requests carry no system prompt and no temperature knob (mirrors the original vlm.v1 proto); both providers pin temperature: 0.0.