GLM 5.3 Flash

external

Would run on our hardware, but we do not operate it. Measured via OpenRouter.

88 Logic Index · via OpenRouter
– tokens per second here
– no tier here

Fact sheet

Developer
Z.AI
Size
321B parameters
Context
1M
Input
Text, Image, Video
License
MIT
Hosting
not hosted here

Call

Measured externally: models that would run on our hardware but that we do not operate. We measured them via OpenRouter with the same test, as a look beyond our own stack. Index and tokens are comparable, time is not, and each provider chooses its own quantization.

By test area

Reasoning chains
87
Logic vs. world knowledge
100
Logic grid puzzles
77

Consistency: same answer in every run for 97% of tasks.

More models