GLM 5.2 Performance Benchmarks
5 hours ago
- #Open Weights
- #Performance Metrics
- #AI Model Analysis
- GLM-5.2 (max) is a leading open-weight model with high intelligence (score 51 vs. average 24), notable speed (112 tokens/sec), but high pricing ($1.40/M input tokens, $4.40/M output tokens).
- It is a reasoning model with 1M token context window, text-only input/output, 753B total parameters (40B active), and verbose output (140M tokens for evaluation).
- Released under MIT license, it supports self-hosting and is compared to similar-sized open-weight models in performance metrics.