⚡ Zen LM
Models

zen-eco

Efficient 4B dense model balancing capability and cost for general-purpose tasks.

zen-eco

Efficient

A 4B dense transformer that balances capability and efficiency. Designed for high-throughput production workloads where cost matters but quality cannot be compromised. Achieves 33K tokens/sec with 2--8GB RAM.

Specifications

PropertyValue
Model IDzen-eco
Parameters4B
ArchitectureDense
Context Window32K tokens
Throughput33K tokens/sec
Memory2--8 GB RAM
HuggingFacezenlm/zen-eco-4b

Capabilities

  • General-purpose text generation
  • Summarization and extraction
  • Classification and labeling
  • High-throughput batch processing
  • Cost-efficient production deployment
  • Instruction following

Usage

HuggingFace

pip install transformers torch
from transformers import AutoModelForCausalLM, AutoTokenizer

tokenizer = AutoTokenizer.from_pretrained("zenlm/zen-eco-4b")
model = AutoModelForCausalLM.from_pretrained("zenlm/zen-eco-4b")

inputs = tokenizer("Summarize the key points:", return_tensors="pt")
outputs = model.generate(**inputs, max_new_tokens=256)
print(tokenizer.decode(outputs[0], skip_special_tokens=True))

API

from hanzoai import Hanzo

client = Hanzo(api_key="hk-your-api-key")

response = client.chat.completions.create(
    model="zen-eco",
    messages=[{"role": "user", "content": "Summarize the key benefits of edge computing."}],
)
print(response.choices[0].message.content)

See Also

  • zen-nano -- 0.6B ultra-compact edge model
  • zen -- 8-32B standard foundation model
  • zen3-nano -- 4B edge-optimized variant

On this page