model roundup

Opus 4.7

3 items · started 2026-06-17 · closed 2026-06-21

  1. Hi everyone, I'm dealing with an incredibly stressful situation right now and wanted to see if anyone else has successfully gotten this resolved. I just got hit with over $130 in surprise "On-Demand" usage charges.

  2. Exciting news: GLM-5.2 (Max) ranks #2 in Code Arena: Frontend, with +29pt over Claude Opus 4.7 (Thinking) and only behind Fable 5! GLM-5.2 is the best open model vs Kimi-K2.6 and Minimax-M3 by a large margin.

  3. I have context documents in both the project level and within the session level, which provide rules and expectations for the output. These are based on developing succesful outputs from previous sessions and even has examples.

← all threads