Instructions to use timpal0l/dolphin-2.9-llama3-8b-flashback with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use timpal0l/dolphin-2.9-llama3-8b-flashback with Transformers:

# Use a pipeline as a high-level helper
from transformers import pipeline

pipe = pipeline("text-generation", model="timpal0l/dolphin-2.9-llama3-8b-flashback")

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("timpal0l/dolphin-2.9-llama3-8b-flashback")
model = AutoModelForCausalLM.from_pretrained("timpal0l/dolphin-2.9-llama3-8b-flashback")

Notebooks
Google Colab
Kaggle
Local Apps

vLLM

How to use timpal0l/dolphin-2.9-llama3-8b-flashback with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "timpal0l/dolphin-2.9-llama3-8b-flashback"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "timpal0l/dolphin-2.9-llama3-8b-flashback",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/timpal0l/dolphin-2.9-llama3-8b-flashback

SGLang

How to use timpal0l/dolphin-2.9-llama3-8b-flashback with SGLang:

Install from pip and serve model

# Install SGLang from pip:
pip install sglang
# Start the SGLang server:
python3 -m sglang.launch_server \
    --model-path "timpal0l/dolphin-2.9-llama3-8b-flashback" \
    --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": "timpal0l/dolphin-2.9-llama3-8b-flashback",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker images

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 "timpal0l/dolphin-2.9-llama3-8b-flashback" \
        --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": "timpal0l/dolphin-2.9-llama3-8b-flashback",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use timpal0l/dolphin-2.9-llama3-8b-flashback with Docker Model Runner:
```
docker model run hf.co/timpal0l/dolphin-2.9-llama3-8b-flashback
```
Browse Quantizations to use this model in llama.cpp, Ollama, LM Studio, or any compatible app.

How to use:

from transformers import pipeline

pipe = pipeline(
    task="text-generation",
    model="dolphin-2.9-llama3-8b-flashback",
    device_map="auto"
)

text = "Vad är meningen med livet?"

prompt = f"""
<|im_start|>system
Du är en smart AI assistant som svarar på frågor.<|im_end|>
<|im_start|>user
{text}<|im_end|>
<|im_start|>assistant
"""

pipe(prompt, max_length=512, do_sample=False)

>>> "Meningen med livet är subjektiv och varierar från person till person.
Vissa ser det som att uppnå sina mål,
medan andra fokuserar på att hjälpa andra och göra världen ett bättre ställe.
Det är viktigt att hitta en mening som passar just dig och ditt liv"

config.yaml

models:
  - model: timpal0l/Llama-3-8B-flashback-v1
    # No parameters necessary for base model
  - model: cognitivecomputations/dolphin-2.9-llama3-8b
    parameters:
      density: 0.53
      weight: 0.6
merge_method: dare_ties
base_model: timpal0l/Llama-3-8B-flashback-v1
parameters:
  int8_mask: true
dtype: bfloat16

Downloads last month: 6

Safetensors

Model size

8B params

Tensor type

BF16

Model tree for timpal0l/dolphin-2.9-llama3-8b-flashback

dphn/dolphin-2.9-llama3-8b

timpal0l/Llama-3-8B-flashback-v1

Merge model

this model

Quantizations

2 models