Instructions to use T3Q-LLM/T3Q-LLM-solar10.8-sft-v1.0 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use T3Q-LLM/T3Q-LLM-solar10.8-sft-v1.0 with Transformers:

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

pipe = pipeline("text-generation", model="T3Q-LLM/T3Q-LLM-solar10.8-sft-v1.0")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# pip install -U transformers accelerate
# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("T3Q-LLM/T3Q-LLM-solar10.8-sft-v1.0")
model = AutoModelForCausalLM.from_pretrained("T3Q-LLM/T3Q-LLM-solar10.8-sft-v1.0", device_map="auto")
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=256)
print(tokenizer.decode(outputs[0][inputs["input_ids"].shape[-1]:]))

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use T3Q-LLM/T3Q-LLM-solar10.8-sft-v1.0 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "T3Q-LLM/T3Q-LLM-solar10.8-sft-v1.0"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "T3Q-LLM/T3Q-LLM-solar10.8-sft-v1.0",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/T3Q-LLM/T3Q-LLM-solar10.8-sft-v1.0

SGLang

How to use T3Q-LLM/T3Q-LLM-solar10.8-sft-v1.0 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 "T3Q-LLM/T3Q-LLM-solar10.8-sft-v1.0" \
    --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": "T3Q-LLM/T3Q-LLM-solar10.8-sft-v1.0",
		"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 "T3Q-LLM/T3Q-LLM-solar10.8-sft-v1.0" \
        --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": "T3Q-LLM/T3Q-LLM-solar10.8-sft-v1.0",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use T3Q-LLM/T3Q-LLM-solar10.8-sft-v1.0 with Docker Model Runner:
```
docker model run hf.co/T3Q-LLM/T3Q-LLM-solar10.8-sft-v1.0
```
Browse Quantizations to use this model in llama.cpp, Ollama, LM Studio, or any compatible app.

T3Q-LLM-solar10.8-sft-v1.0

This model is a version of yanolja/EEVE-Korean-Instruct-10.8B-v1.0 that has been fine-tuned with SFT.

Model Developers Chihoon Lee(chihoonlee10), T3Q

hf (pretrained=T3Q-LLM/T3Q-LLM-solar10.8-sft-v1.0), limit: None, provide_description: False, num_fewshot: 0, batch_size: None

Task	Version	Metric	Value		Stderr
kobest_boolq	0	acc	0.9288	±	0.0069
		macro_f1	0.9286	±	0.0069
kobest_copa	0	acc	0.7440	±	0.0138
		macro_f1	0.7434	±	0.0138
kobest_hellaswag	0	acc	0.4880	±	0.0224
		acc_norm	0.5600	±	0.0222
		macro_f1	0.4854	±	0.0224
kobest_sentineg	0	acc	0.8589	±	0.0175
		macro_f1	0.8589	±	0.0175

hf (pretrained=yanolja/EEVE-Korean-Instruct-10.8B-v1.0), limit: None, provide_description: False, num_fewshot: 0, batch_size: None

Task	Version	Metric	Value		Stderr
kobest_boolq	0	acc	0.9188	±	0.0073
		macro_f1	0.9185	±	0.0073
kobest_copa	0	acc	0.7520	±	0.0137
		macro_f1	0.7516	±	0.0136
kobest_hellaswag	0	acc	0.4840	±	0.0224
		acc_norm	0.5580	±	0.0222
		macro_f1	0.4804	±	0.0223
kobest_sentineg	0	acc	0.8514	±	0.0179
		macro_f1	0.8508	±	0.0180

Downloads last month: 30

Safetensors

Model size

11B params

Tensor type

BF16

Model tree for T3Q-LLM/T3Q-LLM-solar10.8-sft-v1.0

Quantizations

3 models

T3Q-LLM
/

T3Q-LLM-solar10.8-sft-v1.0

T3Q-LLM-solar10.8-sft-v1.0

This model is a version of yanolja/EEVE-Korean-Instruct-10.8B-v1.0 that has been fine-tuned with SFT.

Model Developers Chihoon Lee(chihoonlee10), T3Q

Model tree for T3Q-LLM/T3Q-LLM-solar10.8-sft-v1.0

Dataset used to train T3Q-LLM/T3Q-LLM-solar10.8-sft-v1.0

Space using T3Q-LLM/T3Q-LLM-solar10.8-sft-v1.0 1