Instructions to use KYUNGYONG/EXAONE-3.5-32B-Instruct-abliterated-3bit with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use KYUNGYONG/EXAONE-3.5-32B-Instruct-abliterated-3bit with Transformers:

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

pipe = pipeline("text-generation", model="KYUNGYONG/EXAONE-3.5-32B-Instruct-abliterated-3bit", trust_remote_code=True)
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoModelForCausalLM
model = AutoModelForCausalLM.from_pretrained("KYUNGYONG/EXAONE-3.5-32B-Instruct-abliterated-3bit", trust_remote_code=True, dtype="auto")

MLX

How to use KYUNGYONG/EXAONE-3.5-32B-Instruct-abliterated-3bit 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("KYUNGYONG/EXAONE-3.5-32B-Instruct-abliterated-3bit")

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 KYUNGYONG/EXAONE-3.5-32B-Instruct-abliterated-3bit with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "KYUNGYONG/EXAONE-3.5-32B-Instruct-abliterated-3bit"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "KYUNGYONG/EXAONE-3.5-32B-Instruct-abliterated-3bit",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/KYUNGYONG/EXAONE-3.5-32B-Instruct-abliterated-3bit

SGLang

How to use KYUNGYONG/EXAONE-3.5-32B-Instruct-abliterated-3bit 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 "KYUNGYONG/EXAONE-3.5-32B-Instruct-abliterated-3bit" \
    --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": "KYUNGYONG/EXAONE-3.5-32B-Instruct-abliterated-3bit",
		"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 "KYUNGYONG/EXAONE-3.5-32B-Instruct-abliterated-3bit" \
        --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": "KYUNGYONG/EXAONE-3.5-32B-Instruct-abliterated-3bit",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

MLX LM

How to use KYUNGYONG/EXAONE-3.5-32B-Instruct-abliterated-3bit with MLX LM:

Generate or start a chat session

# Install MLX LM
uv tool install mlx-lm
# Interactive chat REPL
mlx_lm.chat --model "KYUNGYONG/EXAONE-3.5-32B-Instruct-abliterated-3bit"

Run an OpenAI-compatible server

# Install MLX LM
uv tool install mlx-lm
# Start the server
mlx_lm.server --model "KYUNGYONG/EXAONE-3.5-32B-Instruct-abliterated-3bit"
# Calling the OpenAI-compatible server with curl
curl -X POST "http://localhost:8000/v1/chat/completions" \
   -H "Content-Type: application/json" \
   --data '{
     "model": "KYUNGYONG/EXAONE-3.5-32B-Instruct-abliterated-3bit",
     "messages": [
       {"role": "user", "content": "Hello"}
     ]
   }'

Docker Model Runner
How to use KYUNGYONG/EXAONE-3.5-32B-Instruct-abliterated-3bit with Docker Model Runner:
```
docker model run hf.co/KYUNGYONG/EXAONE-3.5-32B-Instruct-abliterated-3bit
```

EXAONE-3.5-32B-Instruct-abliterated-3bit

File size: 1,293 Bytes

{
    "activation_function": "silu",
    "architectures": [
        "ExaoneForCausalLM"
    ],
    "attention_dropout": 0.0,
    "auto_map": {
        "AutoConfig": "configuration_exaone.ExaoneConfig",
        "AutoModelForCausalLM": "modeling_exaone.ExaoneForCausalLM",
        "AutoModelForSequenceClassification": "modeling_exaone.ExaoneForSequenceClassification"
    },
    "bos_token_id": 1,
    "embed_dropout": 0.0,
    "eos_token_id": 361,
    "head_dim": 128,
    "hidden_size": 5120,
    "initializer_range": 0.02,
    "intermediate_size": 27392,
    "layer_norm_epsilon": 1e-05,
    "max_position_embeddings": 32768,
    "model_type": "exaone",
    "num_attention_heads": 40,
    "num_key_value_heads": 8,
    "num_layers": 64,
    "pad_token_id": 0,
    "quantization": {
        "group_size": 64,
        "bits": 3
    },
    "quantization_config": {
        "group_size": 64,
        "bits": 3
    },
    "rope_scaling": {
        "factor": 8.0,
        "high_freq_factor": 4.0,
        "low_freq_factor": 1.0,
        "original_max_position_embeddings": 8192,
        "rope_type": "llama3"
    },
    "rope_theta": 1000000.0,
    "tie_word_embeddings": false,
    "torch_dtype": "bfloat16",
    "transformers_version": "4.43.0",
    "use_cache": true,
    "vocab_size": 102400
}