llm 0.36 (2026-09-22) adds gpt-6-sol/gpt-6-luna, lets plugins set supports_conversation=False (raises ConversationNotSupported; first used by llm-typesafe), and wraps reasoning traces in <details> in llm logs Markdown.

Key Takeaways

  • ✓Built-in gpt-6-sol and gpt-6-luna (#1702)
  • ✓Plugins may set supports_conversation=False → ConversationNotSupported (#1692)
  • ✓First consumer: llm-typesafe for Jev/TypeSafe models
  • ✓llm logs wraps reasoning in HTML details (#1701)
  • ✓Upgrade: pip install -U llm==0.36
🔬

In-Depth Technical Analysis

Background

CLI model catalogs lag new GPT-6 IDs, and single-turn decision models break under chat history. llm 0.36 adds Sol/Luna and an explicit supports_conversation flag.

Architecture

Built-in gpt-6-sol and gpt-6-luna. Plugins may set supports_conversation = False → ConversationNotSupported on history; llm chat rejects early. First plugin: llm-typesafe. Reasoning traces in llm logs Markdown wrap in HTML details. Assorted async/logging/schema fixes.

Benchmarks

No SWE scores; verify via OpenAI model docs and issues #1692/#1702.

Get started

pip install -U 'llm==0.36'; see the 0.36 release and docs.