Zen 5
Next-generation agentic models with native chain-of-thought.
zen5
TBANext-generation agentic frontier model trained on 10B+ tokens of real-world tool use, multi-step reasoning, and production workflows. 1M+ token context with native chain-of-thought.
zen5-pro
TBAHigh-throughput agentic model for demanding production workloads. Trained on real-world development patterns with deep chain-of-thought reasoning.
zen5-max
TBAMaximum context agentic model for document-scale analysis. Trained on 10B+ tokens of real-world workflows with extended chain-of-thought.
zen5-ultra
TBADeepest reasoning model in the Zen family. Multi-pass chain-of-thought with self-verification.
zen5-mini
TBAEfficient agentic model delivering zen5-class intelligence at a fraction of the cost.
Zen 4
Latest generation production models with MoDE architecture.
zen4-max
N/AMost capable model for complex reasoning, analysis, and agentic tasks. 1M token context window.
zen4.1
N/AHigh-performance 1M context model for long-document analysis, large codebase reasoning, and agentic workflows. Best balance of intelligence and cost at million-token scale.
zen4
744B (40B active)Flagship MoE model for complex reasoning and multi-domain tasks.
zen4-ultra
744B (40B active)Maximum reasoning capability with extended chain-of-thought on MoE architecture.
zen4-pro
80B (3B active)Efficient MoE model for demanding workloads with strong reasoning at production-grade cost.
zen4-thinking
80B (3B active)Dedicated reasoning model with explicit chain-of-thought capabilities.
zen4-mini
N/AUltra-fast lightweight model optimized for speed and cost efficiency. Ideal for free tier.
Code
Specialized models for code generation, review, and debugging.
zen4-coder
480B (35B active)Code-specialized MoE model for generation, review, debugging, and agentic programming.
zen4-coder-flash
30B (3B active)Lightweight code model optimized for speed and inline completions.
zen4-coder-pro
480BFull-precision BF16 code model for maximum accuracy on complex codebases.
zen-coder
32BBaseline code model for generation and completions.
zen-coder-flash
7BFast code model for inline completions and suggestions.
zen-code
14BLegacy code model (superseded by Zen4 Coder series).
Zen 3
Previous generation API models — language, vision, multimodal, and safety.
zen3-omni
~200BMultimodal model supporting text, vision, audio, and structured output.
zen3-vl
30B (3B active)Vision-language model for image understanding and visual reasoning.
zen3-nano
8BUltra-lightweight model for edge deployment and low-latency tasks. Available on free tier.
zen3-guard
4BContent safety classifier for moderation and guardrails. 9 safety categories, 119 languages.
Embedding & Retrieval
Text embeddings and search reranking via API.
zen3-embedding
3072 dimensionsHigh-quality text embeddings for RAG, search, and classification.
zen3-embedding-medium
4BBalanced embedding model for cost-effective retrieval workloads.
zen3-embedding-small
0.6BLightweight embedding model for high-throughput, low-cost applications.
zen3-reranker
8BHigh-quality reranker for improving retrieval accuracy in RAG pipelines.
zen3-reranker-medium
4BBalanced reranker for cost-effective retrieval quality improvement.
zen3-reranker-small
0.6BLightweight reranker for high-throughput reranking at minimal cost.
zen-embedding
3072 dimensionsFoundation embedding model for search and retrieval.
zen-reranker
568MCross-encoder reranker for search result quality.
Image Generation
Text-to-image generation via API.
zen3-image
N/ABest general-purpose image generation.
zen3-image-max
N/AMaximum quality image generation for professional creative work.
zen3-image-dev
N/ADevelopment model for experimentation and iteration.
zen3-image-fast
N/AFastest image model for real-time generation.
zen3-image-sdxl
N/AHigh-resolution image generation at 1024px.
zen3-image-playground
N/AAesthetic model for artistic image generation.
zen3-image-ssd
1BFastest diffusion model for real-time generation.
zen3-image-jp
N/AJapanese-specialized image generation model.
Audio & Speech
Speech-to-text, text-to-speech, and streaming ASR.
zen3-audio
1.5BBest quality speech-to-text transcription. 100+ languages.
zen3-audio-fast
809MFastest speech-to-text transcription for high-throughput workloads.
zen3-asr
N/AReal-time streaming speech recognition for live transcription and voice agents.
zen3-asr-v1
N/AFirst-generation streaming ASR for legacy compatibility.
zen3-tts
82MHigh-quality text-to-speech with natural prosody. 40+ voices, 8 languages.
zen3-tts-hd
N/AMaximum fidelity text-to-speech for broadcast-quality audio production.
zen3-tts-fast
82MLow-latency text-to-speech for real-time voice agents and interactive applications.
Foundation
General-purpose open-weight models from 0.6B to 235B parameters.
zen-nano
0.6BUltra-lightweight LLM for edge and mobile deployment.
zen-eco
4BEfficient 4B model for general-purpose tasks.
zen
8–32BStandard model available in 8B and 32B variants.
zen-pro
32BProfessional-grade 32B dense model for demanding workloads.
zen-max
235B (22B active)High-capability MoE model with 235B parameters.
zen-next
TBDNext-generation preview model with cutting-edge capabilities.
Vision (Open Weights)
Vision-language and multimodal open-weight models.
Safety
Content moderation and safety guardrail models.
Agents
Agent-optimized models for tool use and planning.
Capabilities Matrix
Each model specializes in different modalities and tasks
| Model | Text | Image | Video | Audio | 3D | Code | Agents |
|---|---|---|---|---|---|---|---|
| zen5 | ✓ | — | — | — | — | ✓ | ✓ |
| zen4 | ✓ | — | — | — | — | ✓ | ✓ |
| zen4-max | ✓ | — | — | — | — | ✓ | ✓ |
| zen4-ultra | ✓ | — | — | — | — | ✓ | ✓ |
| zen4-coder | ✓ | — | — | — | — | ✓ | ✓ |
| zen3-omni | ✓ | ✓ | — | ✓ | — | — | — |
| zen3-vl | ✓ | ✓ | — | — | — | — | — |
| zen3-nano | ✓ | — | — | — | — | ✓ | — |
| zen3-guard | ✓ | — | — | — | — | — | — |
| zen3-image | ✓ | ✓ | — | — | — | — | — |
| zen3-audio | ✓ | — | — | ✓ | — | — | — |
| zen3-tts | ✓ | — | — | ✓ | — | — | — |
| zen3-embedding | ✓ | — | — | — | — | — | — |
| zen3-reranker | ✓ | — | — | — | — | — | — |
Infrastructure
Production-ready tools for training and deploying Zen models
Zoo Engine
High-performance cloud inference — 60+ architectures, CUDA/Metal, OpenAI-compatible API
Zoo Edge
On-device AI inference — run models locally on any device, browser, or embedded system
Zen Gym
Unified training platform for all Zen models with LoRA, QLoRA, GRPO, and more
Zoo MCP
Model Context Protocol for AI context management and tool use
Quick Start
Get started with any Zen model in seconds
# Install and run any model
pip install transformers torch
# Use directly
from transformers import AutoModelForCausalLM, AutoTokenizer
model = AutoModelForCausalLM.from_pretrained("zenlm/zen-eco-4b-instruct")
tokenizer = AutoTokenizer.from_pretrained("zenlm/zen-eco-4b-instruct")
# Or use via Zoo Cloud API
from zooai import Zoo
client = Zoo(api_key="zk-your-api-key")
response = client.chat.completions.create(
model="zen-eco-4b-instruct",
messages=[{"role": "user", "content": "Hello!"}]
)