AI Models - Documentation

Aymo gives you access to the best AI models in one place. It brings together leading models from Anthropic, OpenAI, Google, xAI, DeepSeek, Meta, Qwen, Mistral, Kimi, MiniMax, MiMo, and Z.ai.

Choose the best model for your task and switch any time.

Whether you're writing, coding, researching with live web search, or analyzing images, there's a model built for it.

Use this page to explore what each model is best at and the capabilities it supports.

Every model is grouped by provider, with quick badges so you can find the right fit at a glance.

Capabilities you'll see across Aymo's models:

Reasoning: works through problems step by step for more accurate answers
Live reasoning: streams its thinking in real time so you can follow along
Web search: pulls in current information from the live web
Vision: reads and understands images, screenshots, and diagrams
Audio input: accepts audio as input
File context: works with uploaded files and very long documents

How to choose a model

Coding and building: Claude Opus 4.8, GPT-5.5, DeepSeek V4 Pro, GLM-5.2, Qwen3 Coder Next, and Qwen3 Coder 30B are tuned for coding and building.
Writing and content: Claude Sonnet 4.6, GPT-5.5, Gemini 2.5 Flash-Lite, Qwen3.7-Max, and Llama 4 Scout shine at writing.
Analysis and research: Gemini 3.1 Pro, DeepSeek R1, MiniMax M2.7, Perplexity Sonar, and Grok 4.3 are built for deep analysis.
Real-time info: Need current answers? Choose a model with web search, like Perplexity Sonar, Grok 4.3, the Gemini 3.x line, GPT-5.4 Nano/Mini, or GPT-5.5.
Images and screenshots: For vision, reach for any Claude model, the Gemini or GPT line, Llama 4, MiMo-V2.5, Qwen3-VL, Qwen3.6, or MiniMax-M3.
Show-your-work answers: Want to watch the model think? Pick one with live reasoning, such as Claude models, the GPT-5 line, DeepSeek R1, DeepSeek V4 Flash, MiMo-V2.5-Pro, Kimi K2.5/K2.6, MiniMax M2.7, or Qwen3.7-Max.
Long documents: For file context, the Claude and Gemini lines handle very long documents with ease.

Anthropic — Claude

Claude models are known for strong reasoning, deep understanding of context, and high-quality writing. Every Claude model supports vision and file analysis, making them useful for coding, research, and document-based workflows.

Claude Opus 4.8

Anthropic's smartest model. Best for tackling difficult questions, writing high-quality content, and complex coding. Choose it when you need the most intelligent and detailed responses.

Best at: Coding, Analysis, Writing
Capabilities: Live reasoning, Vision, File context

Claude Sonnet 4.6

A powerful, well-rounded model great for coding, detailed writing, and long conversations. The go-to choice for most everyday tasks on Aymo.

Best at: Coding, Analysis, Writing
Capabilities: Live reasoning, Vision, File context

Claude Haiku 4.5

The fastest Claude model on Aymo. Great for quick questions, short tasks, and back-and-forth conversations where speed matters.

Best at: Coding, Analysis, Writing
Capabilities: Live reasoning, Vision, File context

OpenAI — GPT

OpenAI's GPT models range from lightweight options for everyday tasks to GPT-5.5, the flagship model for advanced work. Most GPT models support vision and live reasoning, and some also include web search.

GPT-5.5

OpenAI's most capable model on Aymo. Best for complex coding, detailed analysis, and high-quality writing that requires deeper reasoning.

Best at: Coding, Analysis, Writing
Capabilities: Live reasoning, Web search, Vision

GPT-5.4 Mini

A compact model with built-in web search. Well-suited for coding, research, and conversations that benefit from current information.

Best at: Coding, Analysis
Capabilities: Live reasoning, Web search, Vision

GPT-5 Mini

A balanced model that combines speed and quality. A good choice for coding, writing, and everyday tasks that require more depth.

Best at: Coding, Writing, Analysis
Capabilities: Live reasoning, Vision

GPT-5.4 Nano

A fast model with web search built in. Useful for quick research, summaries, and tasks that require current information.

Best at: Analysis, Writing, Coding
Capabilities: Live reasoning, Web search, Vision

GPT-5 Nano

Blazing fast and very affordable. Great for quick lookups, summarizing content, and simple everyday questions.

Best at: Analysis, Writing, Coding
Capabilities: Live reasoning, Vision

GPT-4o Mini

A fast, budget-friendly model that handles everyday writing, answering questions, and light analysis with ease.

Best at: Writing, Analysis, Coding
Capabilities: Vision

Google — Gemini and Gemma

Google's Gemini line combines reasoning with the broadest input support on Aymo, including vision, audio input, and file context. Gemini 3.x models also support web search for up-to-date answers.

Gemini 3.1 Pro

Google's most advanced model on Aymo. Best for in-depth reasoning, long documents, and complex tasks that need the full power of Gemini.

Best at: Analysis, Coding, Writing
Capabilities: Reasoning, Web search, Vision, Audio input, File context

Gemini 3.5 Flash

A cutting-edge Gemini model with web search and image understanding. Great for coding, research, and complex conversations that need current information.

Best at: Coding, Writing, Analysis
Capabilities: Reasoning, Web search, Vision, Audio input, File context

Gemini 3.1 Flash-Lite

An affordable Gemini 3 model with web search and strong instruction following. Ideal for everyday conversations, content creation, and frequent use.

Best at: Writing, Analysis, Coding
Capabilities: Reasoning, Web search, Vision, Audio input, File context

Gemini 2.5 Pro

Google's top reasoning model. Excellent for complex math, coding challenges, deep analysis, and working with large documents.

Best at: Coding, Analysis, Writing
Capabilities: Reasoning, Vision, Audio input, File context

Gemini 2.5 Flash

A smart and fast model that can handle complex questions while still being quick and affordable for everyday use.

Best at: Coding, Analysis, Writing
Capabilities: Reasoning, Vision, Audio input, File context

Gemini 2.5 Flash-Lite

Google's lightest and fastest model. Perfect for quick replies, simple questions, and high-volume chat at the lowest cost.

Best at: Writing, Analysis, Coding
Capabilities: Reasoning, Vision, Audio input, File context

Gemma 4

A versatile open model that understands images, writes code, and supports more than 140 languages. Great for everyday tasks across a wide range of use cases.

Best at: Coding, Analysis, Writing
Capabilities: Reasoning, Vision, File context

xAI — Grok

Grok combines reasoning, web search, and vision in a single model. It's designed for research, current information, and conversations that benefit from real-time web access.

Grok 4.3

xAI's flagship model with built-in reasoning and web search. Great for deep analysis, research-heavy conversations, and finding up-to-date information.

Best at: Analysis, Coding, Writing
Capabilities: Reasoning, Web search, Vision

Perplexity

Perplexity focuses on search-backed answers and current information. Its models combine AI reasoning with live web search, making them useful for research, fact-checking, and staying up to date.

Perplexity Sonar

Answers backed by live web search. Great for staying current, quick research, and real-time fact-checking.

Best at: Analysis
Capabilities: Web search, Live reasoning

DeepSeek

DeepSeek models are known for strong reasoning, coding capabilities, and transparent thinking. Several models support live reasoning, making them a popular choice for complex analysis and development tasks.

DeepSeek V4 Pro

DeepSeek's flagship model. Best for complex coding, deep analysis, and challenging problems that require strong reasoning.

Best at: Coding, Analysis, Writing
Capabilities: Reasoning

DeepSeek V4 Flash

A fast DeepSeek model designed for high-volume conversations and long documents. A good choice when speed and responsiveness matter most.

Best at: Writing, Coding, Analysis
Capabilities: Live reasoning

DeepSeek R1

A top-tier reasoning model that shows its thinking process as it works. Excellent for complex coding challenges and thorough analysis.

Best at: Coding, Analysis, Writing
Capabilities: Live reasoning

DeepSeek V3.2

A balanced model that combines strong reasoning and coding performance at a lower cost. Ideal for everyday technical and analytical tasks.

Best at: Writing, Coding, Analysis
Capabilities: Live reasoning

Qwen

Qwen offers a wide range of models, from coding-focused assistants to vision-capable reasoning models. The lineup includes options for coding, analysis, image understanding, and everyday tasks.

Qwen3.7-Max

Qwen's most capable model on Aymo. Delivers strong performance across coding, analysis, and writing, with visible step-by-step reasoning.

Best at: Writing, Coding, Analysis
Capabilities: Live reasoning

Qwen3 VL 235B A22B Thinking

A powerful vision model that analyzes images and thinks through complex problems step by step. Great for visual reasoning and detailed analysis.

Best at: Analysis, Coding, Writing
Capabilities: Reasoning, Vision

Qwen3.6

A solid all-rounder with image understanding and strong reasoning. Well-suited for complex coding, analysis, and technical conversations.

Best at: Coding, Analysis, Writing
Capabilities: Reasoning, Vision

Qwen3.6-27B

A capable model with strong coding, STEM, and image understanding skills. A good choice for both technical and creative tasks.

Best at: Coding, Analysis, Writing
Capabilities: Reasoning, Vision

Qwen3 Coder Next

A coding-focused model built specifically for software development. Handles large codebases, debugging, and complex programming tasks.

Best at: Coding, Analysis
Capabilities: Reasoning

Qwen3 Next 80B A3B Thinking

A reasoning-focused model that works through problems step by step before answering. Great for deep analysis and multi-step tasks.

Best at: Analysis, Writing, Coding
Capabilities: Reasoning

Qwen3 Coder 30B A3B Instruct

A lightweight coding model built for code generation, technical instructions, and development workflows. Fast, focused, and efficient.

Best at: Coding, Analysis
Capabilities: Text-only

MiniMax

MiniMax models focus on reasoning, long conversations, and complex problem-solving. The lineup includes multimodal capabilities and reasoning-focused models for coding, analysis, and research.

MiniMax-M3

A multimodal model that understands both images and text. Built for long conversations, complex analysis, and demanding coding tasks.

Best at: Coding, Writing, Analysis
Capabilities: Reasoning, Vision

MiniMax M2.7

A reasoning-focused model designed for complex analysis and problem-solving. A strong choice for coding, research, and multi-step tasks.

Best at: Analysis, Coding, Writing
Capabilities: Live reasoning

Z.ai — GLM

GLM models are designed for coding, reasoning, and complex development workflows. The lineup includes both advanced models for demanding tasks and lightweight options for everyday use.

GLM-5.2

Z.ai's latest flagship model on Aymo. Built for long-horizon coding, complex engineering tasks, and agent workflows that require strong reasoning and large context windows.

Best at: Coding, Analysis, Writing

Capabilities: Reasoning

GLM-5.1

A powerful model for coding, development, and long-running tasks. Well-suited for complex workflows that require planning, execution, and problem-solving.

Best at: Coding, Analysis, Writing
Capabilities: Reasoning

GLM-4.7-Flash

A fast and responsive model built for everyday tasks and high-volume usage. A good choice when you need a balance of speed and capability.

Best at: Coding, Analysis, Writing
Capabilities: Reasoning

Kimi

Kimi models are built for long conversations and complex problem-solving. They perform well on coding, analysis, and tasks that require maintaining context over extended sessions.

Kimi K2.6

Stays sharp and reliable across very long sessions. Perfect for complex, ongoing conversations that need consistent quality from start to finish.

Best at: Coding, Analysis, Writing
Capabilities: Live reasoning

Kimi K2.5

A dependable model with strong coding skills and solid performance across long, involved conversations. A proven and reliable choice.

Best at: Coding, Analysis, Writing
Capabilities: Live reasoning

MiMo

MiMo models are built for reasoning, multimodal understanding, and complex problem-solving. The lineup includes models for long, multi-step tasks as well as image and text-based workflows.

MiMo-V2.5-Pro

Built for long, multi-step tasks. This model thinks through big problems from start to finish, showing its reasoning along the way.

Best at: Coding, Analysis, Writing
Capabilities: Live reasoning

MiMo-V2.5

A versatile model that understands both text and images, making it great for visual analysis, coding help, and mixed-media conversations.

Best at: Coding, Analysis, Writing
Capabilities: Reasoning, Vision

Meta — Llama

Llama models are designed for coding, analysis, and long-context conversations. They support image understanding and are well-suited for research, content creation, and everyday tasks.

Llama 4 Maverick

A powerful Meta model that punches well above its price. Delivers top-tier quality for coding, analysis, and writing at a fraction of the usual cost.

Best at: Coding, Analysis, Writing
Capabilities: Vision

Llama 4 Scout Instruct

Meta's model is built for long, detailed conversations. Can read images and handle very large documents, making it great for in-depth research and analysis.

Best at: Writing, Coding, Analysis
Capabilities: Vision

Mistral

Mistral models are designed for efficient text generation, multilingual communication, and instruction following. They are well-suited for writing, analysis, and everyday language tasks.

Mistral Nemo

A compact multilingual model that follows instructions well and works across many languages. Great for international use cases.

Best at: Writing, Analysis, Coding
Capabilities: Text-only

Frequently asked questions

Which AI models can I use on Aymo?

Aymo gives you access to models from Anthropic (Claude), OpenAI (GPT), Google (Gemini & Gemma), xAI (Grok), Perplexity, DeepSeek, Qwen, MiniMax, Z.ai (GLM), Kimi, MiMo, Meta (Llama), and Mistral. You can switch between models at any time and choose the one that best fits your task.

Which models are best for coding?

For coding, strong choices include Claude Opus 4.8, GPT-5.5, DeepSeek V4 Pro, GLM-5.2, Qwen3 Coder Next, and Qwen3 Coder 30B.

Which models can search the web?

Models with web search include Perplexity Sonar, Grok 4.3, GPT-5.5, GPT-5.4 Nano, GPT-5.4 Mini, and the Gemini 3.x line, including Gemini 3.1 Flash-Lite, Gemini 3.5 Flash, and Gemini 3.1 Pro.

Which models can understand images?

Vision-capable models include all Claude models, the Gemini and GPT families, Llama 4 Maverick and Scout, MiMo-V2.5, Qwen3 VL 235B A22B Thinking, Qwen3.6, Grok 4.3, and MiniMax-M3.

Which models show their reasoning?

Models that stream their step-by-step thinking (live reasoning) include the Claude models, the GPT-5 line, DeepSeek R1, DeepSeek V4 Flash, DeepSeek V3.2, MiMo-V2.5-Pro, Kimi K2.5 and K2.6, MiniMax M2.7, and Qwen3.7-Max.

Which models handle long documents?

For long documents and uploaded files, the Claude, Gemini, and Gemma model families are the best options. These models are designed to work with large amounts of context and long documents.

Can I switch between models on Aymo?

Yes. Aymo is a multi-model app, so you can choose the model that best fits each task and switch at any time without setting up separate accounts or API keys.

Glossary

Best at: The kinds of tasks a model performs best, such as coding, analysis, or writing.
Reasoning: The model works through a problem step by step before answering, helping improve accuracy on complex tasks.
Live reasoning: The model streams its thinking in real time so you can follow along as it works.
Web search: The model can pull in current information from the live web.
Vision: The model can read and analyze images, screenshots, and diagrams.
Audio input: The model can accept audio.
File context: The model can work with the contents of uploaded files and long documents.
Text-only: The model works with text only and does not support images, audio, or web search.