Back to Models
Zhipu AIOpen Source

GLM-4.6V-Flash 9B

GLM-4.6V-Flash 9B is a multimodal foundation model developed by Zhipu AI. It has approximately 90.0B parameters and supports a 128K context window.

Parameters

90.0B

Context Window

128K

License

MIT

Release Date

2025-12-08

API Pricing

API pricing for this model is not yet available

Strengths

  • Advanced multimodal processing capabilities
  • Long context understanding of 128K tokens
  • Available under MIT license

Weaknesses

  • High computational resource requirements
  • Lack of optimization for specific domains
  • Need for further inference speed improvement

Use Cases

  • Multimodal analysis
  • Large-scale document processing
  • Open-source AI development

Deep Analysis

Architecture

VLM (9B total)

Lightweight vision-language model

Context Window

128K tokens

Multimodal context

API Pricing

Free

0 yuan for calls

Release Date

December 9, 2025

License

Open weights, commercial OK

Modalities

Text + Image + Video

30 high-res images per round

Strengths

  • Completely free API access
  • 9B parameters - lightweight and efficient
  • 128K multimodal context
  • Native Function Call in visual model
  • Open weights with commercial license
  • SOTA visual understanding at its scale

Weaknesses

  • Smaller model may miss complex reasoning
  • Limited to 9B parameters
  • Chinese model ecosystem

Competitor Comparison

ModelArenaSWEGPQAPrice
GLM-4.6V 106B---1 yuan/1M in
Qwen-VL-Plus---Paid
InternVL2-8B---Free (open)

GLM-4.6V-Flash 9B is Zhipu AI's free, lightweight vision-language model. Released December 2025, it offers 128K multimodal context, native Function Call capability, and SOTA visual understanding accuracy at 9B scale, completely free for commercial use.

Analysis generated: 2026-05-24