← Back to Blog
New model

GPT-5.6 Terra and GPT-5.6 Luna API: now on FastMetal

FastMetal

OpenAI's GPT-5.6 Terra (gpt-5.6-terra) and GPT-5.6 Luna (gpt-5.6-luna) are available on FastMetal. Together with the flagship GPT-5.6 Sol, already in the catalog, that puts all three GPT-5.6 tiers behind one API key and one endpoint.

Which tier to pick

  • Terra sits between Sol and Luna. It suits everyday coding, reasoning and agent tasks where Sol is more model than the job needs.
  • Luna is the fast, lower-cost tier, for chat, classification, light agent workflows and anything high-volume or latency-bound.
  • Sol stays the choice for the hardest reasoning work.

Both new models have a context window of about 1.05M tokens and up to 128K output tokens. On FastMetal they take text and image input; file (PDF) input is refused on these routes.

Calling it

curl https://api.fastmetal.ai/v1/chat/completions \
  -H "Authorization: Bearer $FASTMETAL_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "gpt-5.6-terra",
    "messages": [{"role": "user", "content": "Hello"}]
  }'

For lighter work, change model to gpt-5.6-luna. If you use the OpenAI SDK, point base_url at https://api.fastmetal.ai/v1 and the rest of your code stays as it is. To try either model without code, pick it in chat.

How they compare

For head-to-heads with real answers to the same prompts, see GPT-5.6 Terra vs Claude Sonnet 5, Grok 4.6 vs GPT-5.6 Terra and Gemini 3.7 Flash vs GPT-5.6 Luna.

Pricing

Both bill per million input and output tokens, with cache reads and writes billed per million tokens too, from a prepaid balance. Luna is the cheaper of the two. Current rates are on the GPT-5.6 Terra and GPT-5.6 Luna model pages and on pricing. The full list of models is in the catalog.

Try GPT-5.6 Terra right now

GPT-5.6 Terra is available on FastMetal through one API key. Start in the browser, or call it from the OpenAI SDK.