GLM-4.6

Zhipu AI🇨🇳 China
active

355B MIT-licensed model with 200K context and strong coding capability.

Context window200K tokens
Input / 1M tokens$0.3
Output / 1M tokens$0.9

Version History

glm-4-6-launchmajor

GLM-4.6 launches.

Benchmark Scores

Full leaderboard →
93.9%
AIME 2025
1425.0 elo
Arena Elo
63.2%
GPQA
66.1%
Hallucination (Knowledge)
9.5%
Hallucination Rate
56.1%
LiveCodeBench
78.4%
MMLU-Pro
66.0 tokens_per_sec
Speed (tok/s)
68.0%
SWE-bench Verified