Instructions to use shism/calme-3.2-instruct-78b-Q8-mlx with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use shism/calme-3.2-instruct-78b-Q8-mlx with Transformers:

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

pipe = pipeline("text-generation", model="shism/calme-3.2-instruct-78b-Q8-mlx")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("shism/calme-3.2-instruct-78b-Q8-mlx")
model = AutoModelForCausalLM.from_pretrained("shism/calme-3.2-instruct-78b-Q8-mlx")
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 shism/calme-3.2-instruct-78b-Q8-mlx 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("shism/calme-3.2-instruct-78b-Q8-mlx")

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
LM Studio

vLLM

How to use shism/calme-3.2-instruct-78b-Q8-mlx with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "shism/calme-3.2-instruct-78b-Q8-mlx"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "shism/calme-3.2-instruct-78b-Q8-mlx",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/shism/calme-3.2-instruct-78b-Q8-mlx

SGLang

How to use shism/calme-3.2-instruct-78b-Q8-mlx 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 "shism/calme-3.2-instruct-78b-Q8-mlx" \
    --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": "shism/calme-3.2-instruct-78b-Q8-mlx",
		"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 "shism/calme-3.2-instruct-78b-Q8-mlx" \
        --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": "shism/calme-3.2-instruct-78b-Q8-mlx",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

MLX LM

How to use shism/calme-3.2-instruct-78b-Q8-mlx with MLX LM:

Generate or start a chat session

# Install MLX LM
uv tool install mlx-lm
# Interactive chat REPL
mlx_lm.chat --model "shism/calme-3.2-instruct-78b-Q8-mlx"

Run an OpenAI-compatible server

# Install MLX LM
uv tool install mlx-lm
# Start the server
mlx_lm.server --model "shism/calme-3.2-instruct-78b-Q8-mlx"
# Calling the OpenAI-compatible server with curl
curl -X POST "http://localhost:8000/v1/chat/completions" \
   -H "Content-Type: application/json" \
   --data '{
     "model": "shism/calme-3.2-instruct-78b-Q8-mlx",
     "messages": [
       {"role": "user", "content": "Hello"}
     ]
   }'

Docker Model Runner
How to use shism/calme-3.2-instruct-78b-Q8-mlx with Docker Model Runner:
```
docker model run hf.co/shism/calme-3.2-instruct-78b-Q8-mlx
```

shism/calme-3.2-instruct-78b-Q8-mlx

The Model shism/calme-3.2-instruct-78b-Q8-mlx was converted to MLX format from MaziyarPanahi/calme-3.2-instruct-78b using mlx-lm version 0.20.5.

Use with mlx

pip install mlx-lm

from mlx_lm import load, generate

model, tokenizer = load("shism/calme-3.2-instruct-78b-Q8-mlx")

prompt="hello"

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

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

Downloads last month: 2

Safetensors

Model size

22B params

Tensor type

F16

U32

MLX

Hardware compatibility

8-bit

Model tree for shism/calme-3.2-instruct-78b-Q8-mlx

Base model

MaziyarPanahi/calme-3.2-instruct-78b

Quantized

(10)

this model

Evaluation results

strict accuracy on IFEval (0-Shot)
Open LLM Leaderboard

80.630
normalized accuracy on BBH (3-Shot)
Open LLM Leaderboard

62.610
exact match on MATH Lvl 5 (4-Shot)
Open LLM Leaderboard

39.950
acc_norm on GPQA (0-shot)
Open LLM Leaderboard

20.360
acc_norm on MuSR (0-shot)
Open LLM Leaderboard

38.530
accuracy on MMLU-PRO (5-shot)
test set Open LLM Leaderboard

70.030