Instructions to use MarkrAI/RAG-KO-Mixtral-7Bx2-v1.15 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use MarkrAI/RAG-KO-Mixtral-7Bx2-v1.15 with Transformers:

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

pipe = pipeline("text-generation", model="MarkrAI/RAG-KO-Mixtral-7Bx2-v1.15")

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("MarkrAI/RAG-KO-Mixtral-7Bx2-v1.15")
model = AutoModelForCausalLM.from_pretrained("MarkrAI/RAG-KO-Mixtral-7Bx2-v1.15")

Notebooks
Google Colab
Kaggle
Local Apps

vLLM

How to use MarkrAI/RAG-KO-Mixtral-7Bx2-v1.15 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "MarkrAI/RAG-KO-Mixtral-7Bx2-v1.15"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "MarkrAI/RAG-KO-Mixtral-7Bx2-v1.15",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/MarkrAI/RAG-KO-Mixtral-7Bx2-v1.15

SGLang

How to use MarkrAI/RAG-KO-Mixtral-7Bx2-v1.15 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 "MarkrAI/RAG-KO-Mixtral-7Bx2-v1.15" \
    --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": "MarkrAI/RAG-KO-Mixtral-7Bx2-v1.15",
		"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 "MarkrAI/RAG-KO-Mixtral-7Bx2-v1.15" \
        --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": "MarkrAI/RAG-KO-Mixtral-7Bx2-v1.15",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use MarkrAI/RAG-KO-Mixtral-7Bx2-v1.15 with Docker Model Runner:
```
docker model run hf.co/MarkrAI/RAG-KO-Mixtral-7Bx2-v1.15
```
Browse Quantizations to use this model in llama.cpp, Ollama, LM Studio, or any compatible app.

MarkrAI/RAG-KO-Mixtral-7Bx2-v1.15

Model Details

Model Developers

MarkrAI - AI Researchers

Base Model

DopeorNope/Ko-Mixtral-v1.3-MoE-7Bx2.

Instruction tuning Method

Using QLoRA.

4-bit quantization
Lora_r: 64
Lora_alpha: 64
Lora_dropout: 0.05
Lora_target_modules: [embed_tokens, q_proj, k_proj, v_proj, o_proj, gate, w1, w2, w3, lm_head]

Hyperparameters

Epoch: 3
Batch size: 64
Learning_rate: 1e-5
Learning scheduler: linear
Warmup_ratio: 0.06

Datasets

Private datasets: HumanF-MarkrAI/Korean-RAG-ver2

Aihub datasets 활용하여서 제작함.

Implmentation Code

### KO-Platypus
from transformers import AutoModelForCausalLM, AutoTokenizer
import torch

repo = "MarkrAI/RAG-KO-Mixtral-7Bx2-v1.15"
OpenOrca = AutoModelForCausalLM.from_pretrained(
        repo,
        return_dict=True,
        torch_dtype=torch.float16,
        device_map='auto'
)
OpenOrca_tokenizer = AutoTokenizer.from_pretrained(repo)

Model Benchmark

Coming soon...

Downloads last month: 5

Safetensors

Model size

13B params

Tensor type

F16

Model tree for MarkrAI/RAG-KO-Mixtral-7Bx2-v1.15

Quantizations

1 model

Collection including MarkrAI/RAG-KO-Mixtral-7Bx2-v1.15

State-Of-The-Art Korean-RAG LM

Collection

Markr AI's RAG LLM (based on Ko-Mixtral) • 5 items • Updated Feb 11, 2024 • 2