modelgrep

Qwen: Qwen3 VL 30B A3B Instruct vs Z.ai: GLM 4.5V

QQwen: Qwen3 VL 30B A3B Instructqwen/qwen3-vl-30b-a3b-instructVSZZ.ai: GLM 4.5Vz-ai/glm-4.5v

Qwen: Qwen3 VL 30B A3B Instruct wins on more metrics (8 of 8), but the right pick depends on what you optimize for — see the breakdown below.

MetricQwen: Qwen3 VL 30B A3B InstructZ.ai: GLM 4.5V
Intelligence Index16.0✓15.1
Coding Index14.3✓10.9
GPQA Diamond70%✓68%
Design Arena Elo——
Speed (tokens/sec)45✓19
Latency370ms✓1.0s
Input price /M$0.130✓$0.600
Output price /M$0.520✓$1.80
Context window262K✓66K
CapabilitiesToolsJSONVisionReasoningToolsJSONVision
Qwen: Qwen3 VL 30B A3B Instruct details →Z.ai: GLM 4.5V details →