J.O.S.I.E.-Dev-v4o
Collection
7 items • Updated • 3
How to use Goekdeniz-Guelmez/J.O.S.I.E.v4o-8b-stage1-beta1 with Transformers:
# Use a pipeline as a high-level helper
from transformers import pipeline
pipe = pipeline("text-generation", model="Goekdeniz-Guelmez/J.O.S.I.E.v4o-8b-stage1-beta1") # pip install -U transformers accelerate
# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM
tokenizer = AutoTokenizer.from_pretrained("Goekdeniz-Guelmez/J.O.S.I.E.v4o-8b-stage1-beta1")
model = AutoModelForCausalLM.from_pretrained("Goekdeniz-Guelmez/J.O.S.I.E.v4o-8b-stage1-beta1", device_map="auto")How to use Goekdeniz-Guelmez/J.O.S.I.E.v4o-8b-stage1-beta1 with vLLM:
# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "Goekdeniz-Guelmez/J.O.S.I.E.v4o-8b-stage1-beta1"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
-H "Content-Type: application/json" \
--data '{
"model": "Goekdeniz-Guelmez/J.O.S.I.E.v4o-8b-stage1-beta1",
"prompt": "Once upon a time,",
"max_tokens": 512,
"temperature": 0.5
}'docker model run hf.co/Goekdeniz-Guelmez/J.O.S.I.E.v4o-8b-stage1-beta1
How to use Goekdeniz-Guelmez/J.O.S.I.E.v4o-8b-stage1-beta1 with SGLang:
# Install SGLang from pip:
pip install sglang
# Start the SGLang server:
python3 -m sglang.launch_server \
--model-path "Goekdeniz-Guelmez/J.O.S.I.E.v4o-8b-stage1-beta1" \
--host 0.0.0.0 \
--port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/completions" \
-H "Content-Type: application/json" \
--data '{
"model": "Goekdeniz-Guelmez/J.O.S.I.E.v4o-8b-stage1-beta1",
"prompt": "Once upon a time,",
"max_tokens": 512,
"temperature": 0.5
}'docker run --gpus all \
--shm-size 32g \
-p 30000:30000 \
-v ~/.cache/huggingface:/root/.cache/huggingface \
--env "HF_TOKEN=<secret>" \
--ipc=host \
lmsysorg/sglang:latest \
python3 -m sglang.launch_server \
--model-path "Goekdeniz-Guelmez/J.O.S.I.E.v4o-8b-stage1-beta1" \
--host 0.0.0.0 \
--port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/completions" \
-H "Content-Type: application/json" \
--data '{
"model": "Goekdeniz-Guelmez/J.O.S.I.E.v4o-8b-stage1-beta1",
"prompt": "Once upon a time,",
"max_tokens": 512,
"temperature": 0.5
}'How to use Goekdeniz-Guelmez/J.O.S.I.E.v4o-8b-stage1-beta1 with Docker Model Runner:
docker model run hf.co/Goekdeniz-Guelmez/J.O.S.I.E.v4o-8b-stage1-beta1
This model will be used for the base model in J.O.S.I.E.v4o and furhter trained.
This llama model was trained 2x faster with Unsloth and Huggingface's TRL library.
"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"
"""<|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|>"""
Base model
meta-llama/Meta-Llama-3-8B