You can deploy this model and start earning money today!

RichardErkhov/
Goekdeniz-Guelmez_-_J.O.S.I.E.v4o-8b-stage1-beta1-gguf

Name: Goekdeniz-Guelmez_-_J.O.S.I.E.v4o-8b-stage1-beta1-gguf
Availability: InStock

endpoints_compatible

gguf

template

Views

No views yet

Quantization made by Richard Erkhov.

Github

Discord

Request more models

J.O.S.I.E.v4o-8b-stage1-beta1 - GGUF

Model creator: https://huggingface.co/Goekdeniz-Guelmez/
Original model: https://huggingface.co/Goekdeniz-Guelmez/J.O.S.I.E.v4o-8b-stage1-beta1/

Name	Quant method	Size
J.O.S.I.E.v4o-8b-stage1-beta1.Q2_K.gguf	Q2_K	2.96GB
J.O.S.I.E.v4o-8b-stage1-beta1.IQ3_XS.gguf	IQ3_XS	3.28GB
J.O.S.I.E.v4o-8b-stage1-beta1.IQ3_S.gguf	IQ3_S	3.43GB
J.O.S.I.E.v4o-8b-stage1-beta1.Q3_K_S.gguf	Q3_K_S	3.41GB
J.O.S.I.E.v4o-8b-stage1-beta1.IQ3_M.gguf	IQ3_M	3.52GB
J.O.S.I.E.v4o-8b-stage1-beta1.Q3_K.gguf	Q3_K	3.74GB
J.O.S.I.E.v4o-8b-stage1-beta1.Q3_K_M.gguf	Q3_K_M	3.74GB
J.O.S.I.E.v4o-8b-stage1-beta1.Q3_K_L.gguf	Q3_K_L	4.03GB
J.O.S.I.E.v4o-8b-stage1-beta1.IQ4_XS.gguf	IQ4_XS	4.18GB
J.O.S.I.E.v4o-8b-stage1-beta1.Q4_0.gguf	Q4_0	4.34GB
J.O.S.I.E.v4o-8b-stage1-beta1.IQ4_NL.gguf	IQ4_NL	4.38GB
J.O.S.I.E.v4o-8b-stage1-beta1.Q4_K_S.gguf	Q4_K_S	4.37GB
J.O.S.I.E.v4o-8b-stage1-beta1.Q4_K.gguf	Q4_K	4.58GB
J.O.S.I.E.v4o-8b-stage1-beta1.Q4_K_M.gguf	Q4_K_M	4.58GB
J.O.S.I.E.v4o-8b-stage1-beta1.Q4_1.gguf	Q4_1	4.78GB
J.O.S.I.E.v4o-8b-stage1-beta1.Q5_0.gguf	Q5_0	5.21GB
J.O.S.I.E.v4o-8b-stage1-beta1.Q5_K_S.gguf	Q5_K_S	5.21GB
J.O.S.I.E.v4o-8b-stage1-beta1.Q5_K.gguf	Q5_K	5.34GB
J.O.S.I.E.v4o-8b-stage1-beta1.Q5_K_M.gguf	Q5_K_M	5.34GB
J.O.S.I.E.v4o-8b-stage1-beta1.Q5_1.gguf	Q5_1	5.65GB
J.O.S.I.E.v4o-8b-stage1-beta1.Q6_K.gguf	Q6_K	6.14GB
J.O.S.I.E.v4o-8b-stage1-beta1.Q8_0.gguf	Q8_0	7.95GB

Original model description:

base_model: unsloth/llama-3-8b-bnb-4bit language:

en license: apache-2.0 tags:
text-generation-inference
transformers
unsloth
llama
trl
sft

Still in Beta!!!

This model will be used for the base model in J.O.S.I.E.v4o and furhter trained.

Uploaded model

Developed by: Isaak-Carter
License: apache-2.0
Finetuned from model : Isaak-Carter/JOSIEv4o-8b-stage1-beta1
Dataset used : Isaak-Carter/j.o.s.i.e.v4.0.1o

This llama model was trained 2x faster with Unsloth and Huggingface's TRL library.

json

  "per_device_train_batch_size": 2,
  "gradient_accumulation_steps": 8,
  "max_steps": 300,
  "learning_rate": 2e-4,
  "optim": "adamw_8bit",
  "weight_decay": 0.01,
  "lr_scheduler_type": "cosine"

trained on this Prompt format

txt

"""<|begin_of_text|>system
You are J.O.S.I.E. which is an acronym for "Just an Outstandingly Smart Intelligent Entity", a private and super-intelligent AI assistant, created by Gökdeniz Gülmez.<|end_of_text|>
<|begin_of_text|>main user "Gökdeniz Gülmez"
{{ .Prompt }}<|end_of_text|>
<|begin_of_text|>josie
{{ .Response }}<|end_of_text|>"""

RichardErkhov/Goekdeniz-Guelmez_-_J.O.S.I.E.v4o-8b-stage1-beta1-gguf

Original model description:

Still in Beta!!!

Uploaded model

trained on this Prompt format

RichardErkhov/
Goekdeniz-Guelmez_-_J.O.S.I.E.v4o-8b-stage1-beta1-gguf