Quantization made by Richard Erkhov.

tinyllama_frankenmerge - GGUF

Model creator: https://huggingface.co/bn22/
Original model: https://huggingface.co/bn22/tinyllama_frankenmerge/

Name	Quant method	Size
tinyllama_frankenmerge.Q2_K.gguf	Q2_K	0.55GB
tinyllama_frankenmerge.IQ3_XS.gguf	IQ3_XS	0.61GB
tinyllama_frankenmerge.IQ3_S.gguf	IQ3_S	0.64GB
tinyllama_frankenmerge.Q3_K_S.gguf	Q3_K_S	0.64GB
tinyllama_frankenmerge.IQ3_M.gguf	IQ3_M	0.67GB
tinyllama_frankenmerge.Q3_K.gguf	Q3_K	0.71GB
tinyllama_frankenmerge.Q3_K_M.gguf	Q3_K_M	0.71GB
tinyllama_frankenmerge.Q3_K_L.gguf	Q3_K_L	0.77GB
tinyllama_frankenmerge.IQ4_XS.gguf	IQ4_XS	0.79GB
tinyllama_frankenmerge.Q4_0.gguf	Q4_0	0.82GB
tinyllama_frankenmerge.IQ4_NL.gguf	IQ4_NL	0.83GB
tinyllama_frankenmerge.Q4_K_S.gguf	Q4_K_S	0.83GB
tinyllama_frankenmerge.Q4_K.gguf	Q4_K	0.87GB
tinyllama_frankenmerge.Q4_K_M.gguf	Q4_K_M	0.87GB
tinyllama_frankenmerge.Q4_1.gguf	Q4_1	0.91GB
tinyllama_frankenmerge.Q5_0.gguf	Q5_0	1.0GB
tinyllama_frankenmerge.Q5_K_S.gguf	Q5_K_S	1.0GB
tinyllama_frankenmerge.Q5_K.gguf	Q5_K	1.02GB
tinyllama_frankenmerge.Q5_K_M.gguf	Q5_K_M	1.02GB
tinyllama_frankenmerge.Q5_1.gguf	Q5_1	1.08GB
tinyllama_frankenmerge.Q6_K.gguf	Q6_K	1.18GB
tinyllama_frankenmerge.Q8_0.gguf	Q8_0	1.53GB

Original model description:

license: apache-2.0 tags: - merge - mergekit

tinyllama_frankenmerge

This model is a merge of the following models made with mergekit:

🧩 Configuration

slices:
  - sources:
    - model: TinyLlama/TinyLlama-1.1B-intermediate-step-1195k-token-2.5T
      layer_range: [0, 16]
  - sources:
    - model: TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T
      layer_range: [6, 22]
merge_method: passthrough
dtype: float16