Describe Image Using Vision LLM

Unified dispatcher for image description using vision LLMs. Routes to the appropriate provider (Ollama, OpenAI, or Gemini).

Usage

describe_image(
  image_base64,
  provider = "ollama",
  model = NULL,
  api_key = NULL,
  prompt =
    "Describe this image: charts, diagrams, tables, and text. Extract visible text.",
  timeout = 120
)

Arguments

image_base64

Character string of base64-encoded PNG image

provider

Character: "ollama", "openai", or "gemini"

model

Character: Model name (uses provider default if NULL)

api_key

Character: API key (required for openai/gemini)

prompt

Character: Description prompt

timeout

Numeric: Request timeout in seconds (default: 120)

Value

Character string description, or NULL on failure

See also

Other ai: call_gemini_chat(), call_llm_api(), call_ollama(), call_openai_chat(), check_ollama(), generate_topic_content(), get_api_embeddings(), get_best_embeddings(), get_content_type_prompt(), get_content_type_user_template(), get_recommended_ollama_model(), list_ollama_models(), run_rag_search()