Instructions to use keitokei1994/Llama-3-Umievo-Shizuko-sqlcoder-2x8B with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use keitokei1994/Llama-3-Umievo-Shizuko-sqlcoder-2x8B with Transformers:

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

pipe = pipeline("text-generation", model="keitokei1994/Llama-3-Umievo-Shizuko-sqlcoder-2x8B")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("keitokei1994/Llama-3-Umievo-Shizuko-sqlcoder-2x8B")
model = AutoModelForCausalLM.from_pretrained("keitokei1994/Llama-3-Umievo-Shizuko-sqlcoder-2x8B")
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 keitokei1994/Llama-3-Umievo-Shizuko-sqlcoder-2x8B with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "keitokei1994/Llama-3-Umievo-Shizuko-sqlcoder-2x8B"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "keitokei1994/Llama-3-Umievo-Shizuko-sqlcoder-2x8B",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/keitokei1994/Llama-3-Umievo-Shizuko-sqlcoder-2x8B

SGLang

How to use keitokei1994/Llama-3-Umievo-Shizuko-sqlcoder-2x8B 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 "keitokei1994/Llama-3-Umievo-Shizuko-sqlcoder-2x8B" \
    --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": "keitokei1994/Llama-3-Umievo-Shizuko-sqlcoder-2x8B",
		"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 "keitokei1994/Llama-3-Umievo-Shizuko-sqlcoder-2x8B" \
        --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": "keitokei1994/Llama-3-Umievo-Shizuko-sqlcoder-2x8B",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use keitokei1994/Llama-3-Umievo-Shizuko-sqlcoder-2x8B with Docker Model Runner:
```
docker model run hf.co/keitokei1994/Llama-3-Umievo-Shizuko-sqlcoder-2x8B
```
Browse Quantizations to use this model in llama.cpp, Ollama, LM Studio, or any compatible app.

モデルの説明(English explanation is below.)

このモデルは、MergeKitツールを使用して作成されたMixture of Experts (MoE) 言語モデルです。

gguf版(今後拡充予定)はこちら。

mradermacherさんがgguf形式での量子化を行ってくれました。こちらから確認できます

umiyukiさんが公開しているLlama-3-Umievo-itr014-Shizuko-8b に、SQLデータセットでファインチューニングされたrdefog/llama-3-sqlcoder-8bを合わせることで、日本語能力とSQL生成能力を両立させようとしたMoEモデルです。

モデルの詳細

モデル名: Llama-3-Umievo-Shizuko-sqlcoder-2x8B
モデルアーキテクチャ: Mixture of Experts (MoE)
ベースモデル: rdefog/llama-3-sqlcoder-8b, defog/llama-3-sqlcoder-8b
マージツール: MergeKit

要求スペック

Q4_K_M量子化モデルであれば、RTX3060 12GBでフルロード可能です。

筆者はWSL2やGoogle Colaboratotry Proでの作成後、Llama.cppとLMstudioにて動作確認を行なっています。

CPU: Ryzen 5 3600
GPU: GeForce RTX 3060 12GB
RAM: DDR4-3200 96GB
OS: Windows 10

Model Description

This model is a Mixture of Experts (MoE) language model created using the MergeKit tool. The gguf version (planned to be expanded in the future) can be found here. This MoE model aims to achieve both Japanese language ability and SQL generation capability by combining Llama-3-Umievo-itr014-Shizuko-8b, released by umiyuki, with rdefog/llama-3-sqlcoder-8b, which has been fine-tuned on an SQL dataset.

Model Details

Model Name: Llama-3-Umievo-Shizuko-sqlcoder-2x8B
Model Architecture: Mixture of Experts (MoE)
Base Models: rdefog/llama-3-sqlcoder-8b, defog/llama-3-sqlcoder-8b
Merge Tool: MergeKit

Required Specifications

If using the Q4_K_M quantized model, it can be fully loaded on an RTX 3060 12GB. The author has created the model using WSL2 and Google Colaboratory Pro, and has tested it using Llama.cpp and LMstudio.

CPU: Ryzen 5 3600
GPU: GeForce RTX 3060 12GB
RAM: DDR4-3200 96GB
OS: Windows 10

Downloads last month: 8

Safetensors

Model size

14B params

Tensor type

BF16

Model tree for keitokei1994/Llama-3-Umievo-Shizuko-sqlcoder-2x8B

Quantizations

1 model