Home

Appendix

Application Icon   AI Models

AI models are constantly evolving, with new models appearing, updating, and sometimes disappearing. While a complete compendium of models and capabilities is impossible to create and maintain, here is a list of the commercial models currently supported, listing the size of the context window and supported functions, e.g., vision.

The icons for the functions are as follows: for vision, for tooling, for thinking, and for deep search.

OpenAI:

  • Icon
    gpt-5-nano: 400000 tokens    
  • Icon
    gpt-4.1-nano: 1048576 tokens  
  • Icon
    gpt-5-mini: 400000 tokens    
  • Icon
    gpt-4.1-mini: 1048576 tokens  
  • Icon
    o4-mini: 200000 tokens    
  • Icon
    gpt-5.2: 400000 tokens    
  • Icon
    gpt-4.1: 1048576 tokens  
  • Icon
    o3: 200000 tokens    

Anthropic:

  • Icon
    Claude 4.5 Haiku: 199999 tokens    
  • Icon
    Claude 4.6 Sonnet: 199999 tokens    
  • Icon
    Claude 4.6 Opus: 199999 tokens    

Google:

  • Icon
    gemini-flash-lite: 1048576 tokens    
  • Icon
    gemini-flash: 1048576 tokens    
  • Icon
    gemini-pro: 1048576 tokens    

Mistral:

  • Icon
    mistral-small-latest: 131072 tokens  
  • Icon
    mistral-medium-latest: 131072 tokens  
  • Icon
    mistral-large-latest: 131072 tokens  
  • Icon
    magistral-small-latest: 131072 tokens    
  • Icon
    magistral-medium-latest: 131072 tokens    

Perplexity:

  • Icon
    sonar: 131072 tokens  
  • Icon
    sonar-reasoning-pro: 131072 tokens    
  • Icon
    sonar-deep-research: 131072 tokens  
  • Icon
    sonar-pro: 199999 tokens  

OpenRouter:

  • Icon
    amazon/nova-2-lite-v1: 1000000 tokens    
  • Icon
    amazon/nova-micro-v1: 128000 tokens
  • Icon
    amazon/nova-premier-v1: 1000000 tokens  
  • Icon
    amazon/nova-pro-v1: 300000 tokens  
  • Icon
    anthropic/claude-haiku-4.5: 200000 tokens    
  • Icon
    anthropic/claude-opus-4.6: 1000000 tokens    
  • Icon
    anthropic/claude-sonnet-4.6: 1000000 tokens    
  • Icon
    baidu/ernie-4.5-300b-a47b: 123000 tokens
  • Icon
    baidu/ernie-4.5-vl-424b-a47b: 123000 tokens  
  • Icon
    deepseek/deepseek-v3.2: 163840 tokens  
  • Icon
    deepseek/deepseek-v3.2-speciale: 163840 tokens
  • Icon
    deepseek/deepseek-r1: 64000 tokens  
  • Icon
    google/gemini-2.5-flash-lite: 1048576 tokens    
  • Icon
    google/gemini-3-flash-preview: 1048576 tokens    
  • Icon
    google/gemini-3-pro-preview: 1048576 tokens    
  • Icon
    google/gemma-3-27b-it: 128000 tokens  
  • Icon
    meta-llama/llama-4-maverick: 1048576 tokens  
  • Icon
    meta-llama/llama-4-scout: 327680 tokens  
  • Icon
    mistralai/codestral-2508: 256000 tokens
  • Icon
    mistralai/devstral-2512: 262144 tokens
  • Icon
    mistralai/mistral-large-2512: 262144 tokens  
  • Icon
    mistralai/mistral-medium-3.1: 131072 tokens  
  • Icon
    mistralai/mistral-small-3.2-24b-instruct: 131072 tokens  
  • Icon
    mistralai/mistral-small-creative: 32768 tokens
  • Icon
    mistralai/mistral-saba: 32768 tokens
  • Icon
    moonshotai/kimi-k2.5: 262144 tokens    
  • Icon
    openai/gpt-4.1: 1047576 tokens  
  • Icon
    openai/gpt-4.1-mini: 1047576 tokens  
  • Icon
    openai/gpt-4.1-nano: 1047576 tokens  
  • Icon
    openai/gpt-5-mini: 400000 tokens    
  • Icon
    openai/gpt-5-nano: 400000 tokens    
  • Icon
    openai/gpt-5.1: 400000 tokens    
  • Icon
    openai/gpt-5.1-codex: 400000 tokens    
  • Icon
    openai/gpt-5.1-codex-max: 400000 tokens    
  • Icon
    openai/gpt-5.1-codex-mini: 400000 tokens    
  • Icon
    openai/gpt-5.2: 400000 tokens    
  • Icon
    openai/gpt-5.2-pro: 400000 tokens    
  • Icon
    openai/gpt-5.2-codex: 400000 tokens    
  • Icon
    openai/gpt-oss-120b: 131072 tokens  
  • Icon
    openai/gpt-oss-20b: 131072 tokens  
  • Icon
    openai/o3: 200000 tokens    
  • Icon
    openai/o4-mini: 200000 tokens    
  • Icon
    openai/o4-mini-high: 200000 tokens    
  • Icon
    perplexity/sonar: 127072 tokens  
  • Icon
    perplexity/sonar-deep-research: 128000 tokens  
  • Icon
    perplexity/sonar-pro: 200000 tokens  
  • Icon
    perplexity/sonar-pro-search: 200000 tokens    
  • Icon
    perplexity/sonar-reasoning-pro: 128000 tokens    
  • Icon
    qwen/qwen3-235b-a22b: 131072 tokens  
  • Icon
    qwen/qwen3-235b-a22b-2507: 262144 tokens  
  • Icon
    qwen/qwen3-235b-a22b-thinking-2507: 131072 tokens  
  • Icon
    qwen/qwen3-coder-flash: 1000000 tokens
  • Icon
    qwen/qwen3-coder-next: 262144 tokens
  • Icon
    qwen/qwen3-coder-plus: 1000000 tokens
  • Icon
    qwen/qwen3.5-397b-a17b: 262144 tokens    
  • Icon
    qwen/qwen3.5-plus-02-15: 1000000 tokens    
  • Icon
    x-ai/grok-4: 256000 tokens    
  • Icon
    x-ai/grok-4.1-fast: 2000000 tokens    
  • Icon
    x-ai/grok-code-fast-1: 256000 tokens  
  • Icon
    z-ai/glm-4.6v: 131072 tokens    
  • Icon
    z-ai/glm-4.7-flash: 202752 tokens  
  • Icon
    z-ai/glm-5: 204800 tokens  

Bear in mind, local AI models are not listed here due to the lack of standardization. Depending on your hardware (and your level of patience), you may be running larger models than someone on a less powerful machine. Also, you may have a penchant for a specific AI model provider and want to use their models.