Instructions to use bambuuai/Kai-30B-Instruct-i1-Q4_K_M-GGUF with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use bambuuai/Kai-30B-Instruct-i1-Q4_K_M-GGUF with Transformers:

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

pipe = pipeline("text-generation", model="bambuuai/Kai-30B-Instruct-i1-Q4_K_M-GGUF")

# Load model directly
from transformers import AutoModel
model = AutoModel.from_pretrained("bambuuai/Kai-30B-Instruct-i1-Q4_K_M-GGUF", dtype="auto")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use bambuuai/Kai-30B-Instruct-i1-Q4_K_M-GGUF with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "bambuuai/Kai-30B-Instruct-i1-Q4_K_M-GGUF"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "bambuuai/Kai-30B-Instruct-i1-Q4_K_M-GGUF",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/bambuuai/Kai-30B-Instruct-i1-Q4_K_M-GGUF

SGLang

How to use bambuuai/Kai-30B-Instruct-i1-Q4_K_M-GGUF 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 "bambuuai/Kai-30B-Instruct-i1-Q4_K_M-GGUF" \
    --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": "bambuuai/Kai-30B-Instruct-i1-Q4_K_M-GGUF",
		"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 "bambuuai/Kai-30B-Instruct-i1-Q4_K_M-GGUF" \
        --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": "bambuuai/Kai-30B-Instruct-i1-Q4_K_M-GGUF",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use bambuuai/Kai-30B-Instruct-i1-Q4_K_M-GGUF with Docker Model Runner:
```
docker model run hf.co/bambuuai/Kai-30B-Instruct-i1-Q4_K_M-GGUF
```

Kai-30B-Instruct-i1-Q4_K_M-GGUF / README.md

bambuuai

Upload folder using huggingface_hub

649982c verified 3 months ago

preview code

raw

history blame contribute delete

316 Bytes

metadata

base_model: NoesisLab/Kai-30B-Instruct
language:
  - en
library_name: transformers
license: apache-2.0
bambuuai:
  readme_rev: 1
quantized_by: bambuuai
tags:
  - math
  - reasoning
  - text-generation
  - ads
  - distillation
  - code

About

weighted/imatrix quants of https://huggingface.co/NoesisLab/Kai-30B-Instruct