AI Benchmark Model
Quasar 438B (max, based on GLM-5.2)
Benchmark scores, pricing, speed, and model comparisons for Quasar 438B (max, based on GLM-5.2).
Key scores
Quasar 438B is an artificial intelligence model developed by Multiverse Computing and released on August 10, 2026. The model demonstrates an intelligence index of 43 and a coding index of 61.2. Performance metrics for the model include an output speed of 186.745 tokens per second and a time to first token of 0.583 seconds. Benchmark results for Quasar 438B are recorded as 0.732 on GPQA, 0.187 on HLE, 0.45 on SciCode, and 0.75 on LCR. Pricing for the model is structured at $0.60 per 1 million input tokens and $1.80 per 1 million output tokens, resulting in a blended rate of $0.90 per 1 million tokens.
Benchmark results
4 tracked benchmark rows for Quasar 438B (max, based on GLM-5.2).
LCR Other
76.3% GPQA Reasoning
73.2% SciCode Other
48.1% HLE Other
18.7%