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
| Property | Value |
|---|---|
| Model ID | zen-eco |
| Parameters | 4B |
| Architecture | Dense |
| Context Window | 32K tokens |
| Throughput | 33K tokens/sec |
| Memory | 2--8 GB RAM |
| HuggingFace | zenlm/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 torchfrom 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)