GLM-4.7

Mature GLM model for dependable coding, reasoning, and structured agent tasks

ECI STANDING#87 of 266
SCORE143.46141.57–145.43 CI
FRONTIER LAGLaunched 8.7 months behindScores about what Gemini 2.5 Pro (Mar 2025) did in March 2025.
SCORES LIKE THESE MODELS
Qwen3-235B-A22B-Thinking (Jul 2025)Alibaba · Jul 2025143.88
Qwen 3.6 35B-A3BAlibaba · Apr 2026143.86Qwen 3.6 FlashAlibaba · Apr 2026143.27
Gemini 2.5 Flash (Sep 2025)Google DeepMind · Sep 2025142.99
Full Epoch Capabilities Index ranking

From the lab

Launch announcement · Dec 22, 2025GLM-4.7
RELEASE DATEDec 22, 2025
WEIGHTSOpen weights
CONTEXT WINDOW204.8K
MAX OUTPUT131.1K
REASONINGYes
TOOL CALLINGYes
INPUTStext
OUTPUTStext

Specifications from Models.dev. Availability and limits may differ by provider. Benchmark scores come from Epoch AI under CC BY 4.0.