This model NexVeridian/gpt-oss-120b-3bit was converted to MLX format from openai/gpt-oss-120b using mlx-lm version 0.28.4.
pip install mlx-lm
from mlx_lm import load, generate
model, tokenizer = load("NexVeridian/gpt-oss-120b-3bit")
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)
model-00012-of-00012.safetensors 5.4 GBmodel-00001-of-00010.safetensors 5.4 GBmodel-00001-of-00012.safetensors 5.3 GBmodel-00002-of-00010.safetensors 5.2 GBmodel-00002-of-00012.safetensors 5.2 GBmodel-00005-of-00010.safetensors 5.2 GBmodel-00008-of-00010.safetensors 5.2 GBmodel-00006-of-00010.safetensors 5.2 GBmodel-00009-of-00010.safetensors 5.2 GBmodel-00003-of-00010.safetensors 5.2 GBmodel-00007-of-00010.safetensors 5.2 GBmodel-00004-of-00010.safetensors 5.2 GBmodel-00006-of-00012.safetensors 5.1 GBmodel-00005-of-00012.safetensors 5.1 GBmodel-00009-of-00012.safetensors 5.1 GBmodel-00007-of-00012.safetensors 5.1 GBmodel-00010-of-00012.safetensors 5.1 GBmodel-00011-of-00012.safetensors 5.1 GBmodel-00004-of-00012.safetensors 5.1 GBmodel-00008-of-00012.safetensors 5.1 GBmodel-00003-of-00012.safetensors 5.1 GBmodel-00010-of-00010.safetensors 4.5 GBopenai/gpt-oss-120b