JackCloudman/QwQ-56B-Ghost

QwQ-56B-Ghost is an extended model derived from Qwen-32B-Preview-Jackterated using the merging passthrough technique inspired in models like miqu-1-120b. The result is a ~56B parameter model

Merge Details

Merge Method

This model was merged using the passthrough merge method.

Models Merged

The following models were included in the merge:

  • JackCloudman/QwQ-32B-Preview-jackterated

Configuration

The following YAML configuration was used to produce this model:

dtype: float16
merge_method: passthrough
slices:
  - sources:
      - layer_range: [0, 16]
        model: JackCloudman/QwQ-32B-Preview-jackterated
  - sources:
      - layer_range: [8, 24]
        model: JackCloudman/QwQ-32B-Preview-jackterated
  - sources:
      - layer_range: [16, 32]
        model: JackCloudman/QwQ-32B-Preview-jackterated
  - sources:
      - layer_range: [24, 40]
        model: JackCloudman/QwQ-32B-Preview-jackterated
  - sources:
      - layer_range: [32, 48]
        model: JackCloudman/QwQ-32B-Preview-jackterated
  - sources:
      - layer_range: [40, 56]
        model: JackCloudman/QwQ-32B-Preview-jackterated
  - sources:
      - layer_range: [48, 64]
        model: JackCloudman/QwQ-32B-Preview-jackterated

Credits

Downloads last month
18
Safetensors
Model size
56.2B params
Tensor type
FP16
·
Inference Providers NEW
This model is not currently available via any of the supported third-party Inference Providers, and the model is not deployed on the HF Inference API.

Model tree for JackCloudman/QwQ-56B-Ghost

Base model

Qwen/Qwen2.5-32B
Finetuned
(1)
this model
Quantizations
3 models