Are AI agents shape rotators? In this new benchmark, we let the models play campaign puzzles in Opus Magnum, a puzzle game by @zachtronics.
Ironically, Claude Opus 4.8 performed poorly, being beaten by GPT-5.5, Gemini 3.5 Flash, and GLM 5.2. Claude Fable 5 crushed them all. https://t.co/0TzcFp32B6
GLM-5.2
Open release plus an agent harness aimed squarely at Anthropic; press benchmarked it against Claude Mythos on vulnerability detection. GLM 5.5 reportedly due Aug 2026.
Sources
Official
- 2026-06-13 SCMP: Zhipu releases GLM-5.2 harness — the Anthropic-shaped ambition.
- 2026-06 SCMP: stock rockets on open-source GLM-5.2 — market reaction.
Writing & commentary
tk — to collect
Tweets
tk — fleet pass pending; mirror to archive
Official record
tk — release date, specs, benchmarks, checkpoints
History
tk — context of the world at release; events and entanglements
Impressions
tk — day-of and longitudinal vibes to collect
Records
Full reproductions of the tweets cited on this page — text, images, and verbatim transcriptions of screenshots — kept here against link rot, credited and linked to their originals. Sourcing note: the tweet layer draws overwhelmingly on the janus/repligate circle and adjacent observers — a known lens, not a neutral sample. Sourced from the community archive and the janus corpus. Yours and you’d rather it weren’t here? Open an issue.
Further records
Cited in this model’s dossier but not in the page prose — reproduced so the archive doesn’t depend on editorial selection.
I've gotten a claim this is indeed anything, so: GLM-5.2 reaction thread. What's cooking?
@ollama @mweinbach how many glm 5.2 in/out MToks do you get for $100 ollama max?
A few early GLM 5.2 notebooks.
1/5 https://t.co/u6Li5RKull
Fun fact: the GLM-5.2 weights could fit comfortably in one of these 400-disc DVD carousels from the early 2000s. https://t.co/KdwR0XS5ZZ
Does this answer your question about its positioning @TheZvi ? I told you it's a hardware issue. GLM 5.2 can be served very, very quickly. Databricks won't stop here, neither will their competitors.
Chinese models are *much better* than Chinese silicon. https://t.co/KjOSNOdU6a https://t.co/XHyGaifIVS
@TheZvi Isn't this ostensibly about glm-5.2's successor model rather than 5.2 itself? Thus a more reasonable characterization seems to be highly unlikely rather than obviously false.