Quantization made by Richard Erkhov.

MN-Halide-12b-v1.0 - GGUF

Model creator: https://huggingface.co/Azazelle/
Original model: https://huggingface.co/Azazelle/MN-Halide-12b-v1.0/

Name	Quant method	Size
MN-Halide-12b-v1.0.Q2_K.gguf	Q2_K	4.46GB
MN-Halide-12b-v1.0.IQ3_XS.gguf	IQ3_XS	4.94GB
MN-Halide-12b-v1.0.IQ3_S.gguf	IQ3_S	5.18GB
MN-Halide-12b-v1.0.Q3_K_S.gguf	Q3_K_S	5.15GB
MN-Halide-12b-v1.0.IQ3_M.gguf	IQ3_M	5.33GB
MN-Halide-12b-v1.0.Q3_K.gguf	Q3_K	5.67GB
MN-Halide-12b-v1.0.Q3_K_M.gguf	Q3_K_M	5.67GB
MN-Halide-12b-v1.0.Q3_K_L.gguf	Q3_K_L	6.11GB
MN-Halide-12b-v1.0.IQ4_XS.gguf	IQ4_XS	6.33GB
MN-Halide-12b-v1.0.Q4_0.gguf	Q4_0	6.59GB
MN-Halide-12b-v1.0.IQ4_NL.gguf	IQ4_NL	6.65GB
MN-Halide-12b-v1.0.Q4_K_S.gguf	Q4_K_S	6.63GB
MN-Halide-12b-v1.0.Q4_K.gguf	Q4_K	6.96GB
MN-Halide-12b-v1.0.Q4_K_M.gguf	Q4_K_M	6.96GB
MN-Halide-12b-v1.0.Q4_1.gguf	Q4_1	7.26GB
MN-Halide-12b-v1.0.Q5_0.gguf	Q5_0	7.93GB
MN-Halide-12b-v1.0.Q5_K_S.gguf	Q5_K_S	7.93GB
MN-Halide-12b-v1.0.Q5_K.gguf	Q5_K	8.13GB
MN-Halide-12b-v1.0.Q5_K_M.gguf	Q5_K_M	8.13GB
MN-Halide-12b-v1.0.Q5_1.gguf	Q5_1	8.61GB
MN-Halide-12b-v1.0.Q6_K.gguf	Q6_K	9.37GB
MN-Halide-12b-v1.0.Q8_0.gguf	Q8_0	12.13GB

Original model description:

base_model: - TheDrummer/Rocinante-12B-v1 - jtatman/mistral_nemo_12b_reasoning_psychology_lora - Epiculous/Azure_Dusk-v0.2 - nbeerbower/mistral-nemo-bophades-12B - Epiculous/Crimson_Dawn-v0.2 - nbeerbower/mistral-nemo-wissenschaft-12B - anthracite-org/magnum-v2-12b - TheDrummer/Rocinante-12B-v1.1 - anthracite-org/magnum-v2.5-12b-kto - mpasila/Mistral-freeLiPPA-LoRA-12B - anthracite-org/magnum-v2.5-12b-kto - jeiku/Aura-NeMo-12B - nbeerbower/mistral-nemo-cc-12B - UsernameJustAnother/Nemo-12B-Marlin-v8 - elinas/Chronos-Gold-12B-1.0 - SillyTilly/mistralai_Mistral-Nemo-Base-2407 - nbeerbower/Lyra4-Gutenberg-12B - nbeerbower/mistral-nemo-gutenberg-12B-v4 library_name: transformers tags: - mergekit - merge license: apache-2.0

MN-Halide-12b-v1.0

This is a merge of pre-trained language models created using mergekit.

Merge Details

Merge Method

This model was merged using the Model Stock merge method using SillyTilly/mistralai_Mistral-Nemo-Base-2407 as a base.

Models Merged

The following models were included in the merge:

Configuration

The following YAML configuration was used to produce this model:

base_model: SillyTilly/mistralai_Mistral-Nemo-Base-2407
dtype: float32
merge_method: model_stock
slices:
- sources:
  - layer_range: [0, 40]
    model: nbeerbower/Lyra4-Gutenberg-12B
  - layer_range: [0, 40]
    model: nbeerbower/mistral-nemo-gutenberg-12B-v4
  - layer_range: [0, 40]
    model: elinas/Chronos-Gold-12B-1.0
  - layer_range: [0, 40]
    model: UsernameJustAnother/Nemo-12B-Marlin-v8
  - layer_range: [0, 40]
    model: TheDrummer/Rocinante-12B-v1.1
  - layer_range: [0, 40]
    model: Epiculous/Azure_Dusk-v0.2
  - layer_range: [0, 40]
    model: Epiculous/Crimson_Dawn-v0.2
  - layer_range: [0, 40]
    model: TheDrummer/Rocinante-12B-v1+jtatman/mistral_nemo_12b_reasoning_psychology_lora
  - layer_range: [0, 40]
    model: nbeerbower/mistral-nemo-wissenschaft-12B
  - layer_range: [0, 40]
    model: nbeerbower/mistral-nemo-bophades-12B
  - layer_range: [0, 40]
    model: anthracite-org/magnum-v2.5-12b-kto+mpasila/Mistral-freeLiPPA-LoRA-12B
  - layer_range: [0, 40]
    model: nbeerbower/mistral-nemo-cc-12B
  - layer_range: [0, 40]
    model: anthracite-org/magnum-v2-12b
  - layer_range: [0, 40]
    model: anthracite-org/magnum-v2.5-12b-kto+jeiku/Aura-NeMo-12B
  - layer_range: [0, 40]
    model: SillyTilly/mistralai_Mistral-Nemo-Base-2407
tokenizer_source: unsloth/Mistral-Nemo-Base-2407

Additional thanks to @nicoboss for giving me access to his private supercomputer, enabling me to provide many more quants, at much higher speed, than I would otherwise be able to.