GLM 4 32B 0414 INT8 - caix

Red Hills Media

Run GLM 4 32B 0414 INT8 locally with caix.

Install & run

brew install redhillsmediafl/caix/caix
caix catalog install redhillsmediafl/rhm-glm-4-32b-0414-int8-caix
caix serve --model glm-4-32b-0414-int8-coreai-ctx32768

At a glance

  • Base model: zai-org/GLM-4-32B-0414
  • Context: 32,768 tokens (matches base model)
  • Quantization: int8 Core AI
  • Disk: 30.4 GB
  • Runs on: Apple silicon with caix 0.2.13-beta+

Status

Package Runtime Hardware RAM Model Prompt Output tokens Median tok/s
this package standalone Apple M1 Ultra 64 GB Mac Studio (2022) default decode 128 pending verified result
caix catalog performance-report --model ~/.caix/models/exports/glm-4-32b-0414-int8-coreai-ctx32768 --repo redhillsmediafl/rhm-glm-4-32b-0414-int8-caix

License

MIT

More open-source work: redhillsmediafl.com/open-source.

Downloads last month

-

Downloads are not tracked for this model. How to track
Inference Providers NEW
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support

Model tree for redhillsmediafl/rhm-glm-4-32b-0414-int8-caix

Finetuned
(6)
this model

Collection including redhillsmediafl/rhm-glm-4-32b-0414-int8-caix