# Models — Hanzo AI > Enso is our frontier model family; Zen is our open-weight family. Both run on api.hanzo.ai, and models from other labs answer on the same endpoint with the same key. Live from api.hanzo.ai # Every model, one endpoint Enso is our frontier family, and it answers here only. Zen is Zoo Labs Foundation’s open-weight family — call it here, or download the weights and run it yourself. Every other lab answers on the same address, under the same key. [Get free API key](https://hanzo.ai/signup)[View docs](https://docs.hanzo.ai) ## Enso Our frontier family, trained by Hanzo and served at this address only. [Ensoenso](https://hanzo.ai/models/hanzo/enso)[Enso Flashenso-flash](https://hanzo.ai/models/hanzo/enso-flash)[Enso Ultraenso-ultra](https://hanzo.ai/models/hanzo/enso-ultra) ## Zen The open-weight family from [Zoo Labs Foundation](https://zoo.industries). We serve it first-party — and the weights are published, so you can run it yourself and never call us at all. [Zen5 — Next Generationzen51MNext-generation agentic frontier model with native chain-of-thought.](https://hanzo.ai/models/hanzo/zen5)[Zen5 Pro — Advancedzen5-pro524KHigh-throughput agentic model for demanding production workloads.](https://hanzo.ai/models/hanzo/zen5-pro)[Zen5 Coderzen5-coder](https://hanzo.ai/models/hanzo/zen5-coder)[Zen5 Flashzen5-flash](https://hanzo.ai/models/hanzo/zen5-flash)[Zen VL — Open Visionzen-vl131KOpen-weight vision-language model.](https://hanzo.ai/models/hanzo/zen-vl)[Zen Imagezen-image](https://hanzo.ai/models/hanzo/zen-image)[Zen Embedding — Openzen-embedding8KOpen-weight embedding model.](https://hanzo.ai/models/hanzo/zen-embedding) ## Top models One flagship per lab, so you can find the name you already know. They answer at the same address with the same key, and switching to one is changing a string. [Claude 3 Haikuanthropic/claude-3-haiku200KClaude 3 Haiku is Anthropic's fastest and most compact model for near-instant responsiveness. Quick and accurate targeted performance. See the launch announcement and benchmark results here #multimodaltext](https://hanzo.ai/models/anthropic/claude-3-haiku)[GPT-3.5 Turboopenai/gpt-3.5-turbo16KGPT-3.5 Turbo is OpenAI's fastest model. It can understand and generate natural language or code, and is optimized for chat and traditional completion tasks. Training data up to Sep 2021.text](https://hanzo.ai/models/openai/gpt-3.5-turbo)[Gemini 2.5 Progoogle/gemini-2.5-pro1MGemini 2.5 Pro is Google’s state-of-the-art AI model designed for advanced reasoning, coding, mathematics, and scientific tasks. It employs “thinking” capabilities, enabling it to reason through responses with enhanced accuracy...text](https://hanzo.ai/models/google/gemini-2.5-pro)[DeepSeek V3deepseek/deepseek-chat164KDeepSeek-V3 is the latest model from the DeepSeek team, building upon the instruction following and coding abilities of the previous versions. Pre-trained on nearly 15 trillion tokens, the reported evaluations...text](https://hanzo.ai/models/deepseek/deepseek-chat)[Qwen-Plusqwen/qwen-plus1MQwen-Plus, based on the Qwen2.5 foundation model, is a 131K context model with a balanced performance, speed, and cost combination.text](https://hanzo.ai/models/qwen/qwen-plus)[Grok 4.20x-ai/grok-4.202MGrok 4.20 is a reasoning model from SpaceXAI with industry-leading speed and agentic tool calling capabilities. It combines the lowest hallucination rate on the market with strict prompt adherance, delivering...text](https://hanzo.ai/models/x-ai/grok-4.20)[Muse Glimmer 30Bmeta/muse-glimmer-30b131KMuse Glimmer 30B is a dense, open-weight multimodal model from Meta Superintelligence Labs, distilled from Muse Spark and optimized for autonomous agents on consumer hardware. It is suited for long-horizon...text](https://hanzo.ai/models/meta/muse-glimmer-30b)[Kimi K2 0711moonshotai/kimi-k2131KKimi K2 Instruct is a large-scale Mixture-of-Experts (MoE) language model developed by Moonshot AI, featuring 1 trillion total parameters with 32 billion active per forward pass. It is optimized for...text](https://hanzo.ai/models/moonshotai/kimi-k2) ## Browse by lab Every model on the endpoint, grouped by who trained it. Prices and context windows come from the live catalog, so what you read here is what you will be billed. [Hanzo116 models](https://hanzo.ai/models/hanzo)[OpenAI95 models](https://hanzo.ai/models/openai)[Qwen51 models](https://hanzo.ai/models/qwen)[Google42 models](https://hanzo.ai/models/google)[Anthropic32 models](https://hanzo.ai/models/anthropic)[Mistral20 models](https://hanzo.ai/models/mistralai)[Z.ai16 models](https://hanzo.ai/models/z-ai)[DeepSeek15 models](https://hanzo.ai/models/deepseek)[Meta12 models](https://hanzo.ai/models/meta)[Minimax11 models](https://hanzo.ai/models/minimax)[NVIDIA10 models](https://hanzo.ai/models/nvidia)[Moonshot AI9 models](https://hanzo.ai/models/moonshotai)[ByteDance7 models](https://hanzo.ai/models/bytedance)[xAI7 models](https://hanzo.ai/models/x-ai)[OpenRouter6 models](https://hanzo.ai/models/openrouter)[Tencent6 models](https://hanzo.ai/models/tencent)[Amazon5 models](https://hanzo.ai/models/amazon)[Cohere5 models](https://hanzo.ai/models/cohere)[Perplexity5 models](https://hanzo.ai/models/perplexity)[TThinking Machines5 models](https://hanzo.ai/models/thinkingmachines)[Aion Labs4 models](https://hanzo.ai/models/aion-labs)[Nous Research4 models](https://hanzo.ai/models/nousresearch)[poolside4 models](https://hanzo.ai/models/poolside)[TTheDrummer4 models](https://hanzo.ai/models/thedrummer)[KwaiPilot3 models](https://hanzo.ai/models/kwaipilot)[SSao10K3 models](https://hanzo.ai/models/sao10k)[Arcee AI2 models](https://hanzo.ai/models/arcee-ai)[IBM Granite2 models](https://hanzo.ai/models/ibm-granite)[iinclusionAI2 models](https://hanzo.ai/models/inclusionai)[Microsoft2 models](https://hanzo.ai/models/microsoft)[Morph2 models](https://hanzo.ai/models/morph)[NNex-AGI2 models](https://hanzo.ai/models/nex-agi)[RReka AI2 models](https://hanzo.ai/models/rekaai)[Relace2 models](https://hanzo.ai/models/relace)[SSakana AI2 models](https://hanzo.ai/models/sakana)[StepFun2 models](https://hanzo.ai/models/stepfun)[Upstage2 models](https://hanzo.ai/models/upstage)[Xiaomi2 models](https://hanzo.ai/models/xiaomi)[Allen AI1 model](https://hanzo.ai/models/allenai)[AAnthracite1 model](https://hanzo.ai/models/anthracite-org)[Baidu1 model](https://hanzo.ai/models/baidu)[CCognitive Computations1 model](https://hanzo.ai/models/cognitivecomputations)[ddots-studio1 model](https://hanzo.ai/models/dots-studio)[GGryphe1 model](https://hanzo.ai/models/gryphe)[Inception1 model](https://hanzo.ai/models/inception)[LLiquid AI1 model](https://hanzo.ai/models/liquid)[MMancer1 model](https://hanzo.ai/models/mancer)[Meituan1 model](https://hanzo.ai/models/meituan)[PPerceptron1 model](https://hanzo.ai/models/perceptron)[UUndi951 model](https://hanzo.ai/models/undi95)[WWriter1 model](https://hanzo.ai/models/writer) ## Try it Pick any model and start a thread, signed in with your Hanzo account. Sign in to chat with Enso, Zen, or any model on the endpoint. Pick a model, start a thread, billed to your account and kept private. [Get free API key](https://hanzo.ai/signup)[Open Hanzo Chat](https://hanzo.ai/chat) ## One key, one host Changing model is changing one string. The endpoint takes and returns the chat-completions JSON shape, so an HTTP client already written against that shape works once its base URL points at api.hanzo.ai/v1. typescript ``` curl https://api.hanzo.ai/v1/chat/completions \ -H "Authorization: Bearer $HANZO_API_KEY" \ -H "Content-Type: application/json" \ -d '{ "model": "zen5", "messages": [{ "role": "user", "content": "Hello" }] }' ``` [Get Free API Key](https://hanzo.ai/signup)[Read the Docs](https://docs.hanzo.ai)[Try in Chat](https://hanzo.ai/chat)