Artificial Analysis finds Beam uses ~26k output tokens per Intelligence Index task

1. 9:31 AM
   Artificial Analysis@ArtificialAnlysOfficial

   Beam uses 26k output tokens per Intelligence Index task, mostly reasoning

   AIArtificial Analysis reports that Beam uses about 26k output tokens per Intelligence Index task, with about 19k for reasoning and 7k for answers. Only 26% of its output goes to final answers, the 28th-lowest share among 219 models tested on Intelligence Index v4.3.

   
2. 9:31 AM
   Artificial Analysis@ArtificialAnlysOfficial

   Beam scores 28 on Artificial Analysis Coding Agent Index in Mirror harness

   AIArtificial Analysis says Beam, running in Mirror, Reflection's open-source CLI harness, scores 28 on its Coding Agent Index. That is 15 points behind Claude Code with Qwen3.8 Max (43) and 40 points behind the leaders (68). Only Codex with GPT-6 Astra and Codex with GPT-6.1 Sol use fewer tokens.

   

The trend compares only continuously observed participants, so its scope may be smaller than the current heat total. Hover or tap for hourly heat; use arrow keys with a keyboard.