Together AI Adds GLM-5.3 With Usage Advice
Zain recommends high reasoning effort mode for GLM-5.3 users on the platform.
TLDR
Zain posted a PSA for GLM-5.3 users on X, advising reasoning_effort set to high rather than max. He noted max mode reached 34 percent for 77k tokens while high reached 32 percent for 50k tokens. The post states GLM-5.3 remains quite token efficient. Together AI retweeted the message from its official account. Generated headlines in the packet describe the addition of the model to the platform and note stronger coding performance compared with prior versions.
Combined views
10.1K
2 Sources, first seen ago
Sources
ZA 
Zain@zainhasPSA: For everyone using GLM-5.3 you probably want to use this model in reasoning_effort="high" mode. > "max" - 34% for 77k tokens (not bad!) > "high" - 32% for 50k tokens But "max" is not as unusable as it was for GLM-5.3 Flash. > GLM-5.3 is quite token efficient!…
- likes: 88
- replies: 4
- bookmarks: 25
- reposts: 5