GLM-4.7-Flash-RAM-20GB-MLX

baa-ai/GLM-4.7-Flash-RAM-20GB-MLX

29.9B parametersmlxLicense: other
Tracked by OpenModelStats since October 16, 2026View on Hugging Face ↗
Downloads 30D
29
Total downloads
1,484
Gained 7D
—
Gained 30D
—
Likes
0
Rank · tracked models
—
–this week
Parameters
29.9B
Last source update
Oct 16, 2026

Downloads over time

Rolling recent-window downloads as reported

Likes over time

Rank among tracked models

Lower is better

Not enough tracked history yet for a chart.

OpenModelStats began tracking this model on 2026-10-16. Charts appear as daily observations accumulate.

Overview

GLM-4.7-Flash-RAM-20GB-MLX is an AI model published by baa-ai. OpenModelStats has tracked the model since Oct 16, 2026, most recently observing it Oct 16, 2026.

Published
Apr 6, 2026
Last updated
Oct 16, 2026
Library
mlx
License
other
Task
—
Parameters
29,943,393,920
Spaces
—
Derivative models
—
Velocity 7D/day
—
mlxsafetensorsglm4_moe_litequantizedmixed-precisionglmmoebase_model:zai-org/GLM-4.7-Flashbase_model:quantized:zai-org/GLM-4.7-Flashlicense:other4-bitregion:us

Publisher

baa-aiView publisher statistics →

Derived from zai-org/GLM-4.7-Flash (reported by the source; parent not yet tracked).

Similar models

Models similar to GLM-4.7-Flash-RAM-20GB-MLX
ModelDL 30D
GLM-4.7-Flash-MLX-8bitlmstudio-community169.7K
GLM-4.7-Flash-MLX-6bitlmstudio-community159.4K
GLM-4.7-Flash-4bitmlx-community4,164
GLM-4.7-Flash-MLX-4bitlmstudio-community3,348
Huihui-GLM-4.7-Flash-abliterated-mlx-4bithuihui-ai1,474
GLM-4.7-Flash-8bitmlx-community1,405
baa-ai/GLM-4.7-Flash-RAM-20GB-MLX — Downloads, Growth & Stats | OpenModelStats