nm-testing
/

DeepSeek-Coder-V2-Lite-Instruct-FP8-dynamic

compressed-tensors

Model card Files Files and versions

DeepSeek-Coder-V2-Lite-Instruct-FP8-dynamic

File size: 154 Bytes

7dbc1af

default_stage:
  default_modifiers:
    QuantizationModifier:
      targets: [Linear]
      ignore: [lm_head, 're:.*mlp.gate$']
      scheme: FP8_DYNAMIC