Skip to content

Releases

0.2.0: OpenAI and Anthropic support

Choose a tag to compare

@ian-abu ian-abu released this 19 Nov 10:28

llmburn now tracks OpenAI and Anthropic calls as well as Gemini.

pip install "llmburn[openai]"      # or [anthropic], [gemini], [all]

New

  • OpenAI: chat.completions.create and responses.create, sync and async, streaming included. The SDK's .stream() helpers are tracked too.
  • Anthropic: messages.create and messages.stream, sync and async.
  • One budget across providers. budget_per_user and budget_total count spending from all three SDKs together, so a user who hit their limit on Claude is blocked on GPT too.
  • llmburn.instrument() patches every SDK that's installed and returns their names, e.g. ["gemini", "openai"].
  • Prices for the GPT-6 models and Claude models from Haiku 4.5 up, including cache reads, cache writes (Anthropic 5-minute and 1-hour, OpenAI 30-minute and 12-hour) and long-prompt tiers (GPT-6 above 272k tokens, Claude Haiku 5.5 above 100k).
  • More Gemini models: gemini-4-carbon, gemini-4-hydrogen, gemini-3.6-flash, gemini-3.1-pro-preview-customtools, gemini-robotics-er-2-preview.

Fixed

  • Model names like gemini-2.5-flash-image and gemini-2.5-flash-preview-tts were priced as gemini-2.5-flash. A longer name now only matches a known model when the extra part is a version suffix (-001, -20251001, -2024-08-06, -latest, -preview-05-20). Other models are skipped with a warning instead of being mispriced.
  • gemini-2.5-pro is priced at Gemini 3.8 Flash rates from December 1, 2026, when Google starts billing it that way.

Notes

  • On OpenAI chat streams, llmburn asks for token counts (stream_options={"include_usage": True}) and hides the extra usage-only chunk, so your code sees the same chunks as before. If you asked for usage yourself, you get the chunk as usual.
  • Prices are for the standard tier with global routing. Batch, Flex, Fast and Priority modes and regional surcharges aren't applied.
  • prices.json now uses long_context: {"above": N, ...} in place of above_200k. This only matters if you read the file directly.