Day 0 Support for GPT-6 Astra on LiteLLM

[misbahsy]

Day 0 Support for GPT-6 Astra

What's new:

State of the art on computer use, browsing, software engineering, and cybersecurity - 98% on FrontierMath Tier 4, 99.9% on ARC-AGI 3, and 100% on ExploitBench. A 1,050,000-token context window, 128K max output, and an April 30 2026 knowledge cutoff.

Pricing - $10 / MTok input and $50 / MTok output, cache reads at $1 and cache writes at $12.50. Past 272K input tokens, input and cache rates go to 2x and output to 1.5x for the whole request, not only the tokens above the line, so a long request runs $20 input and $75 output. Batch and Flex take 50% off whichever rate applies, and Fast Mode costs 2x for up to 2.5x the speed.

Effort ladder - reasoning_effort accepts low, medium, high, xhigh, and max.

Get started, no redeploy needed:

Pricing and the request handling land in PR #39607. Reload your pricing in the UI under Models + Endpoints -> Price Data -> Reload Price Data (or POST /reload/model_cost_map as an admin), on any version v1.76.0 or newer.

model_list:
  - model_name: gpt-6-astra
    litellm_params:
      model: openai/gpt-6-astra
      api_key: os.environ/OPENAI_API_KEY

Read the full guide → Day 0 Support: GPT-6 Astra

Misbah Syed
LiteLLM AI Team