Zhipu AI·released 2026-06-161 source
Not yet benchmarked
Claims drawn from cited facts, not live model generation.
Zhipu AI released this model in 2026 at a frontier-scale 753.33 billion parameters, placing it among the largest openly described models of its vintage. That frontier-scale size buys considerable headroom in capability and generality, but it comes at the cost of heavier training and inference expenses rather than the efficiency and lower cost that a more compact parameter count would allow.
4 cited facts
This model is not yet scored on any tracked benchmarks, so no benchmark results are available for it. Consequently it holds no current top score on any tracked benchmark.
2 cited facts
The average gap to the state of the art is unknown, so no frontier-versus-competitive band can be drawn from it. What is known is that this model holds the top score on 0 benchmarks, which means it shows no measured category leadership at all rather than a near-frontier placing.
2 cited facts
2 facts cross-checked across data sources: 2 single-source.
Source facts, citations, and refresh stamp for this record.
Sources: huggingface_models
GLM-5.2 is an AI model developed by Zhipu AI, released 2026-06-16. Its benchmark record below tags every score as independently reproduced, self-reported, or unverified.