Instructions to use dphn/dolphin-2.8-experiment26-7b-preview with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use dphn/dolphin-2.8-experiment26-7b-preview with Transformers:

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

pipe = pipeline("text-generation", model="dphn/dolphin-2.8-experiment26-7b-preview")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("dphn/dolphin-2.8-experiment26-7b-preview")
model = AutoModelForCausalLM.from_pretrained("dphn/dolphin-2.8-experiment26-7b-preview")
messages = [
    {"role": "user", "content": "Who are you?"},
]
inputs = tokenizer.apply_chat_template(
	messages,
	add_generation_prompt=True,
	tokenize=True,
	return_dict=True,
	return_tensors="pt",
).to(model.device)

outputs = model.generate(**inputs, max_new_tokens=40)
print(tokenizer.decode(outputs[0][inputs["input_ids"].shape[-1]:]))

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use dphn/dolphin-2.8-experiment26-7b-preview with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "dphn/dolphin-2.8-experiment26-7b-preview"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "dphn/dolphin-2.8-experiment26-7b-preview",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/dphn/dolphin-2.8-experiment26-7b-preview

SGLang

How to use dphn/dolphin-2.8-experiment26-7b-preview 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 "dphn/dolphin-2.8-experiment26-7b-preview" \
    --host 0.0.0.0 \
    --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "dphn/dolphin-2.8-experiment26-7b-preview",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

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 "dphn/dolphin-2.8-experiment26-7b-preview" \
        --host 0.0.0.0 \
        --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "dphn/dolphin-2.8-experiment26-7b-preview",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use dphn/dolphin-2.8-experiment26-7b-preview with Docker Model Runner:
```
docker model run hf.co/dphn/dolphin-2.8-experiment26-7b-preview
```
Browse Quantizations to use this model in llama.cpp, Ollama, LM Studio, or any compatible app.

1-epoch checkpoint

Please note - this checkpoint release is deprecated in favor of the final release, located here

Open LLM Leaderboard Evaluation Results

Detailed results can be found here

Metric	Value
Avg.	68.60
AI2 Reasoning Challenge (25-Shot)	64.51
HellaSwag (10-Shot)	83.79
MMLU (5-Shot)	63.24
TruthfulQA (0-shot)	54.87
Winogrande (5-shot)	81.61
GSM8k (5-shot)	63.61

Downloads last month: 229

Model tree for dphn/dolphin-2.8-experiment26-7b-preview

Base model

yam-peleg/Experiment26-7B

Finetuned

(10)

this model

Quantizations

2 models

Evaluation results

normalized accuracy on AI2 Reasoning Challenge (25-Shot)
test set Open LLM Leaderboard

64.510
normalized accuracy on HellaSwag (10-Shot)
validation set Open LLM Leaderboard

83.790
accuracy on MMLU (5-Shot)
test set Open LLM Leaderboard

63.240
mc2 on TruthfulQA (0-shot)
validation set Open LLM Leaderboard

54.870
accuracy on Winogrande (5-shot)
validation set Open LLM Leaderboard

81.610
accuracy on GSM8k (5-shot)
test set Open LLM Leaderboard

63.610