Models

OpenAI

GitHub Edit page

OpenAI models

Use OpenAI directly by choosing an openai/... model id.

For the full model/provider matrix, see docs/llm.md.

#GPT-6

Use the documented model ids: openai/gpt-6-astra, openai/gpt-6-sol, openai/gpt-6-luna, or openai/gpt-6.1-sol. Summarize uses the Responses API by default for direct OpenAI text and image requests, including streaming, and for PDF inputs. OPENAI_USE_CHAT_COMPLETIONS=1 still selects Chat Completions for text and images. Reasoning requests omit temperature, which GPT-6 rejects. GPT-6 Sol and Luna preserve temperature when explicitly using --thinking none; Astra and GPT-6.1 Sol require reasoning.

summarize https://summarize.sh/ --model openai/gpt-6-astra --thinking medium --fast
summarize https://summarize.sh/ --model openai/gpt-6-sol --thinking medium
summarize https://summarize.sh/ --model openai/gpt-6-luna --thinking low
summarize https://summarize.sh/ --model openai/gpt-6.1-sol --thinking max

These explicit model ids do not change the default model or existing presets. The model picker and built-in pricing catalog include all four models. Standard input/output prices per million tokens are $10/$50 for Astra, $2/$10 for Sol and GPT-6.1 Sol, and $0.10/$0.50 for Luna. Cost estimates use these standard rates; they do not include cache discounts, cache writes, fast/flex service tiers, or long-context premiums. See OpenAI pricing. All four GPT-6 models support low, medium, high, xhigh, and max. Summarize accepts max only for these documented model ids. An unsupported --thinking none request for Astra or GPT-6.1 Sol fails locally before sending an API request. See OpenAI's GPT-6 guide and model pages for Astra, Sol, Luna, and GPT-6.1 Sol.

#Fast mode

Fast mode is a service tier, not a model id. Prefer flags:

summarize URL --model openai/gpt-5.5 --fast --thinking medium
summarize URL --model openai/gpt-5.4 --service-tier fast --thinking low

Compatibility aliases:

  • --model gpt-fast / --model fast: openai/gpt-5.5 with fast service tier and medium thinking.
  • --model openai/gpt-5.5-fast or --model openai/gpt-5.4-mini-fast: strip the -fast suffix for the API model id and use fast service tier.
  • --model codex-fast: explicit Codex CLI fast preset.

#Env

  • OPENAI_API_KEY (required for openai/... models)
  • OPENAI_USE_CHAT_COMPLETIONS (optional; force chat completions)

#Flags

  • --model openai/<model>
  • --fast
  • --service-tier default|fast|priority|flex
  • --thinking none|low|medium|high|xhigh|max
  • --length short|medium|long|xl|xxl|<chars>
  • This is soft guidance to the model (no hard truncation).
  • --max-output-tokens <count>
  • Hard cap for output tokens (optional).
  • --json (includes prompt + summary in one JSON object)

#Config

{
  "openai": {
    "serviceTier": "fast",
    "thinking": "medium"
  }
}

openai.reasoningEffort is the long-form alias for openai.thinking. Model presets can also set serviceTier, thinking / reasoningEffort, and textVerbosity.

Use "serviceTier": "fast" in summarize config and model presets. OpenAI API requests map that to service_tier="priority"; "flex" and explicit "priority" pass through as-is. Use --service-tier default to clear a configured tier for one run.

#PDF inputs

  • When a PDF is provided and --preprocess auto is used, summarize sends the PDF as a file input via the OpenAI Responses API.
  • Document streaming is disabled for file inputs; non-streaming calls are used instead.