zenlm/zen4-pro-max

Zen4 Pro Max

Zen4 Pro Max is an 80B MoE (3B active) parameter language model from the Zen4 family by Zen LM and Hanzo AI.

The ultimate consumer model with hybrid Gated DeltaNet + Gated Attention + MoE architecture, running at just 3B active parameters. Built on abliterated (uncensored) weights from Qwen3-Next-80B-A3B-Instruct.

Model Details

Property	Value
Parameters	80B total, 3B active
Context	256K tokens
Base	Qwen3-Next-80B-A3B-Instruct (abliterated)
Architecture	Hybrid DeltaNet + MoE, 512 experts
License	Apache 2.0
Family	Zen4
Creator	Zen LM / Hanzo AI

Zen4 Family

Model	Params	Active	Context	HuggingFace
Zen4 Mini	4B	4B	32K	zenlm/zen4-mini
Zen4	8B	8B	32K	zenlm/zen4
Zen4 Pro	14B	14B	32K	zenlm/zen4-pro
Zen4 Max	30B MoE	3B	256K	zenlm/zen4-max
Zen4 Pro Max	80B MoE	3B	256K	zenlm/zen4-pro-max
Zen4 Coder Flash	31B MoE	3B	131K	zenlm/zen4-coder-flash
Zen4 Coder	80B MoE	3B	256K	zenlm/zen4-coder
Zen4 Ultra	1.04T MoE	32B	256K	zenlm/zen4-ultra

Usage

from transformers import AutoModelForCausalLM, AutoTokenizer

model = AutoModelForCausalLM.from_pretrained("zenlm/zen4-pro-max")
tokenizer = AutoTokenizer.from_pretrained("zenlm/zen4-pro-max")

messages = [{"role": "user", "content": "Hello, who are you?"}]
text = tokenizer.apply_chat_template(messages, tokenize=False, add_generation_prompt=True)
inputs = tokenizer(text, return_tensors="pt")
outputs = model.generate(**inputs, max_new_tokens=512)
print(tokenizer.decode(outputs[0], skip_special_tokens=True))

Links

• Zen LM | Hanzo AI
• GitHub
• All Zen4 Models