Skip to content
Skip to workspace
Opening your workspace…

MARKETPLACE / CHAT

AI chat models

Chat with 31 leading AI models, including Gemini 4 Hydrogen, Qwen4 Max, Gemini 4 Carbon and Claude Fable 5.2. Compare context windows, image understanding, reasoning and token prices, and pay only for what you use.

All 31 chat models

Marketplace · Chat

AI chat models, compared

Chat with 31 leading AI models, including Gemini 4 Hydrogen, Qwen4 Max, Gemini 4 Carbon and Claude Fable 5.2. Compare context windows, image understanding, reasoning and token prices, and pay only for what you use.

  • 31 models
  • 10 companies
  • Pay as you go
GEMINIChatQWENChatGEMINIChat

How to choose

Picks based on what each model supports today.

All 31 chat models side by side

Newest first. Prices are starting prices; the exact price for your settings is shown before you send.

All chat models on Flat AI
ModelBest forKey specsPrice
Gemini 4 HydrogenGoogleGoogle's fast, cost-efficient Gemini 4 model for everyday chat, coding and agentic work, with a 1M-token window and…1M context · vision · fast$2.25 in · $11.25 out / 1M
Qwen4 MaxAlibabaAlibaba's new flagship Qwen model for reasoning, coding and agent work, with image and video understanding, web…1M context · image and video · web search$2.14 in · $6.41 out / 1M
Gemini 4 CarbonGoogleGoogle's strongest Gemini 4 model, built for the hardest problems: long engineering, research and analysis tasks,…1M context · vision · deep thinking$3.00 in · $15.00 out / 1M
Claude Fable 5.2AnthropicAnthropic's most capable Claude model for the most demanding technical work: long coding, agentic and research tasks,…1M context · vision · web search$12.00 in · $60.00 out / 1M
GPT-6.1 TerraOpenAIOpenAI's fast mid-tier model for everyday work: writing, coding and understanding screenshots and images, with a…1.1M context · vision · reasoning$1.50 in · $7.50 out / 1M
GPT-6.1 Terra ProOpenAIOpenAI's extended-reasoning version of GPT-6.1 Terra for harder coding and analysis at a mid-tier price, with a…1.1M context · vision · reasoning$1.50 in · $7.50 out / 1M
Kimi K3.1 ProMoonshot AIMoonshot AI's flagship open-weight model, a refined successor to Kimi K3 for long coding, agentic and knowledge work,…1M context · vision · reasoning$3.00 in · $15.00 out / 1M
Gemini 4 ArgonGoogleGoogle's frontier Gemini model for long, demanding work: coding, research and analysis across a 1M-token window, with…1M context · vision · reasoning$3.00 in · $15.00 out / 1M
Muse WonderMetaMeta's frontier reasoning model for complex personal and professional work, with image and video understanding and a…1M context · vision · reasoning$6.00 in · $30.00 out / 1M
Claude Haiku 5.5AnthropicAnthropic's small, fast model for everyday chat, summaries and quick tasks, with a 1M-token context.1M context · vision · reasoning$0.15 in · $0.75 out / 1M
GPT-6.1 SolOpenAIOpenAI's versatile model for coding, analysis and document-heavy work, with a 1M-token context.1.1M context · vision · reasoning$3.00 in · $15.00 out / 1M
GPT-6.1 Sol ProOpenAIOpenAI's extended-reasoning version of GPT-6.1 Sol for complex coding and analysis, with a 1M-token context.1.1M context · vision · reasoning$3.00 in · $15.00 out / 1M
Claude Sonnet 5.5AnthropicAnthropic's versatile model for coding, writing and professional work, with a 1M-token context.1M context · vision · reasoning$3.00 in · $15.00 out / 1M
Qwen3.8 Max PrimeAlibabaAlibaba's top Qwen model on a faster priority tier, for complex reasoning and coding, with image and video…1M context · vision · reasoning$6.00 in · $18.00 out / 1M
Claude Opus 5.5AnthropicAnthropic's flagship model for demanding reasoning, coding and long multi-step work, with a 1M-token context.1M context · vision · reasoning$6.00 in · $30.00 out / 1M
GPT-6 LunaOpenAIOpenAI's fast, low-cost model for everyday chat, quick answers and high-volume tasks, with a 1M-token context.1.1M context · vision · reasoning$0.15 in · $0.75 out / 1M
GPT-6 SolOpenAIOpenAI's cost-efficient high-end model for demanding professional work, analysis and coding, with a 1M-token context.1.1M context · vision · reasoning$3.00 in · $15.00 out / 1M
Grok 4.7xAIxAI's flagship model for coding, research and knowledge work, strong at checking its own work, with a 500K-token…500K context · vision · reasoning$2.40 in · $7.20 out / 1M
MiMo V2.6 FlashXiaomiXiaomi's fast, low-cost open model for everyday chat, with image and video understanding and a 1M-token context.1M context · vision · reasoning$0.21 in · $0.42 out / 1M
MiMo V2.6 ProXiaomiXiaomi's flagship model, with over 1T parameters, for demanding reasoning and coding, with image and video…1.1M context · vision · reasoning$0.65 in · $1.31 out / 1M
DeepSeek V4.1 FlashDeepSeekDeepSeek's fast, very low-cost model for everyday chat and coding, with image understanding and a 1M-token context.1M context · vision · reasoning$0.05 in · $0.44 out / 1M
GPT-6 AstraOpenAIOpenAI's flagship model for the hardest work: deep analysis, research, software engineering and long documents, with…1.1M context · vision · reasoning$15.00 in · $75.00 out / 1M
Claude Fable 5.1AnthropicAnthropic's top-tier model for long code refactors, complex agentic work and knowledge work, with a 1M-token context.1M context · vision · reasoning$15.00 in · $75.00 out / 1M
GLM 5.3 FlashZ.aiFast, low-cost multimodal model for everyday chat, coding and long documents, with a 1.3M-token context.1.3M context · vision · reasoning$0.07 in · $0.21 out / 1M
GLM 5.3Z.aiZ.ai's large reasoning model for complex analysis, software engineering and long-horizon work, with a 1.3M-token…1.3M context · reasoning$2.10 in · $6.60 out / 1M
DeepSeek V4 ProDeepSeekDeepSeek's large mixture-of-experts model for reasoning, maths and coding at a low price, with a 1M-token context.1M context · reasoning$0.44 in · $5.25 out / 1M
Qwen3.7 FlashAlibabaAlibaba's very low-cost vision model for quick chat, image and video understanding, with a 1M-token context.1M context · vision · reasoning$0.05 in · $0.20 out / 1M
Kimi K3Moonshot AIMoonshot AI's 2.8T-parameter open-weight model for complex coding, knowledge work and long tasks, with image and…1M context · vision · reasoning$4.50 in · $22.50 out / 1M
Claude Sonnet 5AnthropicAnthropic's balanced model: frontier-level coding, writing and professional work at a lower price than Opus, with a…1M context · vision · reasoning$3.00 in · $15.00 out / 1M
Gemma 4 31BGoogleGoogle's open 31B model: compact, low-cost and multimodal, with image and video understanding and a 256K-token context.262K context · vision · reasoning$0.14 in · $0.51 out / 1M
Llama 4 MaverickMetaMeta's open multimodal model for fast, inexpensive chat and image understanding, with a 1M-token context.1M context · vision$0.28 in · $0.98 out / 1M

Questions

Which chat model should I use?

It depends on the job. Largest context: GLM 5.3 Flash and GLM 5.3 handle 1.3M tokens. Hard problems: Models that always reason before answering: Gemini 4 Hydrogen, Claude Fable 5.2, GPT-6.1 Terra Pro and Muse Wonder and more. Coding: Gemini 4 Hydrogen, Qwen4 Max, Gemini 4 Carbon and Claude Fable 5.2 and more.

How many chat models does Flat AI offer?

31 chat models from 10 companies, including Gemini 4 Hydrogen, Qwen4 Max, Gemini 4 Carbon, Claude Fable 5.2 and GPT-6.1 Terra. New models are added as they are released.

How does pricing work?

Each model has its own price, charged from your Flat AI balance per token used. There is no subscription, and the price for your settings is shown before you send.

Can I try a model without an account?

You can browse every model page, its settings and examples without an account. To chat, sign in and add funds to your balance.

Can I use these models through an API?

Yes. Every Marketplace model is also available through the Flat AI API with the same balance. See the Developers page for keys and examples.

Marketplace prices are in USD and charged from your Flat AI balance. Models are made by their respective companies.

Where do you want to go?

↑↓ Navigate ↵ Open Esc Close