Instructions to use mlx-community/bigcode-starcoder2-15b-6bit with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use mlx-community/bigcode-starcoder2-15b-6bit with Transformers:

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

pipe = pipeline("text-generation", model="mlx-community/bigcode-starcoder2-15b-6bit")

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("mlx-community/bigcode-starcoder2-15b-6bit")
model = AutoModelForCausalLM.from_pretrained("mlx-community/bigcode-starcoder2-15b-6bit")

MLX

How to use mlx-community/bigcode-starcoder2-15b-6bit with MLX:

# Make sure mlx-lm is installed
# pip install --upgrade mlx-lm
# if on a CUDA device, also pip install mlx[cuda]

# Generate text with mlx-lm
from mlx_lm import load, generate

model, tokenizer = load("mlx-community/bigcode-starcoder2-15b-6bit")

prompt = "Once upon a time in"
text = generate(model, tokenizer, prompt=prompt, verbose=True)

Notebooks
Google Colab
Kaggle
Local Apps
LM Studio

vLLM

How to use mlx-community/bigcode-starcoder2-15b-6bit with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "mlx-community/bigcode-starcoder2-15b-6bit"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "mlx-community/bigcode-starcoder2-15b-6bit",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/mlx-community/bigcode-starcoder2-15b-6bit

SGLang

How to use mlx-community/bigcode-starcoder2-15b-6bit 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 "mlx-community/bigcode-starcoder2-15b-6bit" \
    --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": "mlx-community/bigcode-starcoder2-15b-6bit",
		"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 "mlx-community/bigcode-starcoder2-15b-6bit" \
        --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": "mlx-community/bigcode-starcoder2-15b-6bit",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

MLX LM

How to use mlx-community/bigcode-starcoder2-15b-6bit with MLX LM:

Generate or start a chat session

# Install MLX LM
uv tool install mlx-lm
# Generate some text
mlx_lm.generate --model "mlx-community/bigcode-starcoder2-15b-6bit" --prompt "Once upon a time"

Docker Model Runner
How to use mlx-community/bigcode-starcoder2-15b-6bit with Docker Model Runner:
```
docker model run hf.co/mlx-community/bigcode-starcoder2-15b-6bit
```

bigcode-starcoder2-15b-6bit / README.md

austinbv

Update README.md

cc1eb8b verified over 1 year ago

preview code

raw

history blame contribute delete

2.74 kB

	---
	pipeline_tag: text-generation
	inference:
	parameters:
	temperature: 0.2
	top_p: 0.95
	widget:
	- text: 'def print_hello_world():'
	example_title: Hello world
	group: Python
	datasets:
	- bigcode/the-stack-v2-train
	license: bigcode-openrail-m
	library_name: transformers
	tags:
	- code
	- mlx
	base_model: bigcode/starcoder2-15b
	model-index:
	- name: starcoder2-15b
	results:
	- task:
	type: text-generation
	dataset:
	name: CruxEval-I
	type: cruxeval-i
	metrics:
	- type: pass@1
	value: 48.1
	- task:
	type: text-generation
	dataset:
	name: DS-1000
	type: ds-1000
	metrics:
	- type: pass@1
	value: 33.8
	- task:
	type: text-generation
	dataset:
	name: GSM8K (PAL)
	type: gsm8k-pal
	metrics:
	- type: accuracy
	value: 65.1
	- task:
	type: text-generation
	dataset:
	name: HumanEval+
	type: humanevalplus
	metrics:
	- type: pass@1
	value: 37.8
	- task:
	type: text-generation
	dataset:
	name: HumanEval
	type: humaneval
	metrics:
	- type: pass@1
	value: 46.3
	- task:
	type: text-generation
	dataset:
	name: RepoBench-v1.1
	type: repobench-v1.1
	metrics:
	- type: edit-smiliarity
	value: 74.08
	---

	# mlx-community/bigcode-starcoder2-15b-6bit

	The Model [mlx-community/bigcode-starcoder2-15b-6bit](https://huggingface.co/mlx-community/bigcode-starcoder2-15b-6bit) was
	converted to MLX format from [bigcode/starcoder2-15b](https://huggingface.co/bigcode/starcoder2-15b)
	using mlx-lm version 0.21.1 by [Focused](https://focused.io).

	[![Focused Logo](https://focused.io/images/header-logo.svg "Focused Logo")](https://focused.io)

	## Use with mlx

	```bash
	pip install mlx-lm
	```

	```python
	from mlx_lm import load, generate

	model, tokenizer = load("mlx-community/bigcode-starcoder2-15b-6bit")

	prompt = "hello"

	if tokenizer.chat_template is not None:
	messages = [{"role": "user", "content": prompt}]
	prompt = tokenizer.apply_chat_template(
	messages, add_generation_prompt=True
	)

	response = generate(model, tokenizer, prompt=prompt, verbose=True)
	```

	---

	Focused is a technology company at the forefront of AI-driven development, empowering organizations to unlock the full potential of artificial intelligence. From integrating innovative models into existing systems to building scalable, modern AI infrastructures, we specialize in delivering tailored, incremental solutions that meet you where you are.
	Curious how we can help with your AI next project?
	[Get in Touch](https://focused.io/capabilities/ai-readiness-implementation)

	[![Focused Logo](https://focused.io/images/header-logo.svg "Focused Logo")](https://focused.io)