model roundup

GLM 5.2

5 items · started 2026-07-27 · closed 2026-07-31

  1. Verified inference means cryptographic proof that the exact open model you requested produced your output, not a cheaper or quantized stand-in. Run frontier open models like GLM-5.2, billed by the token.

  2. The danger frontier: low-cost, evasive, abundant malware As part of Incalmo’s mission to make AI safely ubiquitous, we do safety research on the frontier cyber capabilities of models. Recently, to help anti-virus systems stay ahead of the…

  3. Coinbase Switches to Chinese AI Models GLM and Kimi, Cuts AI Spending by 50% - Coinbase has defaulted engineers to GLM 5.2 from Zhipu and Kimi 2.7 from Moonshot AI through its internal LLM gateway, cutting AI spending by nearly 50% [1] - G…

  4. A month ago, GLM-5.2 was released. As part of our day-zero support, we built the fastest API in the world for GLM-5.2, with peak speeds of 280 tokens per second and average speeds around 100 tokens per second.

  5. This wasn’t a photo finish. Sakana: Fugu Ultra controlled the matchup on practical writing and coding tasks, while GLM 5.2 showed flashes of polish in a couple of narrower instruction-following spots.

← all threads