Instructions to use mlx-community/Llama-3.1-Swallow-70B-Instruct-v0.1 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use mlx-community/Llama-3.1-Swallow-70B-Instruct-v0.1 with Transformers:

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

pipe = pipeline("text-generation", model="mlx-community/Llama-3.1-Swallow-70B-Instruct-v0.1")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForMultimodalLM

tokenizer = AutoTokenizer.from_pretrained("mlx-community/Llama-3.1-Swallow-70B-Instruct-v0.1")
model = AutoModelForMultimodalLM.from_pretrained("mlx-community/Llama-3.1-Swallow-70B-Instruct-v0.1")
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]:]))

MLX

How to use mlx-community/Llama-3.1-Swallow-70B-Instruct-v0.1 with MLX:

# Make sure mlx-lm is installed
# pip install --upgrade mlx-lm

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

model, tokenizer = load("mlx-community/Llama-3.1-Swallow-70B-Instruct-v0.1")

prompt = "Write a story about Einstein"
messages = [{"role": "user", "content": prompt}]
prompt = tokenizer.apply_chat_template(
    messages, add_generation_prompt=True
)

text = generate(model, tokenizer, prompt=prompt, verbose=True)

Notebooks
Google Colab
Kaggle
Local Apps Settings
LM Studio

vLLM

How to use mlx-community/Llama-3.1-Swallow-70B-Instruct-v0.1 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "mlx-community/Llama-3.1-Swallow-70B-Instruct-v0.1"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "mlx-community/Llama-3.1-Swallow-70B-Instruct-v0.1",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/mlx-community/Llama-3.1-Swallow-70B-Instruct-v0.1

SGLang

How to use mlx-community/Llama-3.1-Swallow-70B-Instruct-v0.1 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/Llama-3.1-Swallow-70B-Instruct-v0.1" \
    --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": "mlx-community/Llama-3.1-Swallow-70B-Instruct-v0.1",
		"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 "mlx-community/Llama-3.1-Swallow-70B-Instruct-v0.1" \
        --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": "mlx-community/Llama-3.1-Swallow-70B-Instruct-v0.1",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

MLX LM

How to use mlx-community/Llama-3.1-Swallow-70B-Instruct-v0.1 with MLX LM:

Generate or start a chat session

# Install MLX LM
uv tool install mlx-lm
# Interactive chat REPL
mlx_lm.chat --model "mlx-community/Llama-3.1-Swallow-70B-Instruct-v0.1"

Run an OpenAI-compatible server

# Install MLX LM
uv tool install mlx-lm
# Start the server
mlx_lm.server --model "mlx-community/Llama-3.1-Swallow-70B-Instruct-v0.1"
# Calling the OpenAI-compatible server with curl
curl -X POST "http://localhost:8000/v1/chat/completions" \
   -H "Content-Type: application/json" \
   --data '{
     "model": "mlx-community/Llama-3.1-Swallow-70B-Instruct-v0.1",
     "messages": [
       {"role": "user", "content": "Hello"}
     ]
   }'

Docker Model Runner
How to use mlx-community/Llama-3.1-Swallow-70B-Instruct-v0.1 with Docker Model Runner:
```
docker model run hf.co/mlx-community/Llama-3.1-Swallow-70B-Instruct-v0.1
```

License Compatibility

by qiuqiu666 - opened Jun 24, 2025

Discussion

qiuqiu666

Jun 24, 2025

Hi , I’d like to flag a potential license compatibility issue in
mmnga/tokyotech-llm-Llama-3.1-Swallow-70B-Instruct-v0.3-gguf. From what I can tell, this model appears to be a quantized version of tokyotech-llm/Llama-3.1-Swallow-70B-Instruct-v0.3, which is licensed under dual licenses: the LLaMA 3.1 Community License, and the Gemma License from Google .

However, your model is currently published under LLaMA 3.1 only, without mention of the Gemma license, which might be problematic due to the non-transferable and non-permissive nature of the Gemma license terms.

However, the merged model is currently published under the LLaMA 3.3 Community License, which may not be fully compatible with the terms of the Gemma license. This could raise legal and compliance issues regarding redistribution and downstream usage.

⚠️ Key License Compatibility Concerns:

Upstream Model (tokyotech-llm) License:
• Dual-licensed under LLaMA 3.1 + Gemma
• This means **both licenses must be preserved** in any derivative work

Gemma License (by Google):
• Prohibits commercial use
• Restricts redistribution unless terms are passed down unmodified
• Requires explicit license inclusion and attribution
• Includes Google’s Acceptable Use Policy (AUP) — must be propagated downstream

Your Model (mmnga):
• Only includes LLaMA 3.1 Community License
• No mention of Gemma license terms
• No indication that Google’s AUP or usage limitations are preserved

By omitting the Gemma license, this derivative may unintentionally remove critical restrictions on redistribution and commercial usage required by Google — potentially placing downstream users in legal uncertainty.

🔹 Suggestions for Resolving

1. Explicitly state in the model card and/or README that the model is derived from dual-licensed content
2. Add both license texts (LLaMA 3.1 and Gemma) in the repository or model card
3. Include a NOTICE file with:
   • Full attribution to Meta and Google
   • Licensing URLs and copyright statements
4. Clarify usage restrictions:
   • If commercial use is disallowed due to Gemma, the entire model must inherit that restriction
5. If unsure about licensing scope, consider contacting Google and Meta for guidance

Let me know if I misunderstood anything — happy to help clarify further!

Thanks for your attention!

qiuqiu666

Jul 10, 2025

Would love to hear your view on this!

Upload images, audio, and videos by dragging in the text input, pasting, or clicking here.

Tap or paste here to upload images

· Sign up or log in to comment