AI Model Hub

AI Models

Browse GPT-6 Astra, GLM 5.2, Qwen3.8-Max, GPT-5.6 Terra, GPT-5.6 Luna, GPT-5.6 Sol, GPT-5.5, Claude Opus 5, Claude Opus 4.8, GPT Image 2, Grok 4.5, Kimi K3, and DeepSeek V4 model guides for cited-source search, creative content, coding, image generation, and research workflows.

Featured AI model guides

Frontier Reasoning

OpenAI

GPT-6 Astra

A GPT-6 guide for cited-source search, deep reasoning, professional writing, coding help, and complex research workflows.

  • Cited-source search
  • Deep reasoning
  • Coding and analysis
Benchmarks

Not benchmarked

Open model guide
Search + Research

OpenAI

GPT-5.5

A high-capability reasoning model for cited-source answers, long-form writing, coding help, and complex research prompts.

  • Search with cited sources
  • Creative writing
  • Coding and analysis
Benchmarks
Design Arena · Code
GPT-5.5
Elo
1,264
Preference win rate
53.4%
Tournament matches: 0

Reference benchmarks, not measurements of this deployment.

Open model guide
Search + Research

OpenAI

GPT-5.6 Terra

A GPT-5.6 guide for grounded research, structured content, coding support, and complex professional prompts.

  • Grounded research
  • Structured content
  • Coding support
Benchmarks

Not benchmarked

Open model guide
Creative Reasoning

OpenAI

GPT-5.6 Luna

A GPT-5.6 guide for creative content, cited-source answers, planning, coding help, and research workflows.

  • Creative content
  • Cited answers
  • Planning and analysis
Benchmarks

Not benchmarked

Open model guide
Search + Research

OpenAI

GPT-5.6 Sol

A GPT-5.6 guide for cited-source search, deep reasoning, long-form drafting, coding help, and complex research workflows.

  • Cited-source search
  • Deep reasoning
  • Coding and analysis
Benchmarks

Not benchmarked

Open model guide
Agentic Reasoning

Anthropic

Claude Opus 5

An Anthropic model guide for cited-source search, agentic coding, long-form drafting, and complex professional research workflows.

  • Cited search
  • Agentic coding
  • Research workflows
Benchmarks
Design Arena · Code
Claude Opus 5
Elo
1,328
Preference win rate
57.5%
Tournament matches: 0

Reference benchmarks, not measurements of this deployment.

Open model guide
Agentic Reasoning

Anthropic

Claude Opus 4.8

An Anthropic model guide for cited-source search, agentic coding, long-form drafting, and complex professional research workflows.

  • Cited search
  • Agentic coding
  • Research workflows
Benchmarks
Design Arena · Code
Claude Opus 4.8
Elo
1,258
Preference win rate
52.6%
Tournament matches: 0

Reference benchmarks, not measurements of this deployment.

Open model guide
Agentic Reasoning

Anthropic

Claude Fable 5

An Anthropic model guide for cited-source search, advanced drafting, coding help, and professional research workflows.

  • Cited search
  • Advanced drafting
  • Research workflows
Benchmarks

Not benchmarked

Open model guide
Fast Reasoning

Anthropic

Claude Sonnet 5

An Anthropic model guide for cited-source search, fast drafting, coding help, and professional research workflows.

  • Cited search
  • Fast drafting
  • Coding help
Benchmarks
Design Arena · Code
Claude Sonnet 5
Elo
1,281
Preference win rate
52.9%
Tournament matches: 0

Reference benchmarks, not measurements of this deployment.

Open model guide
Fast Reasoning

Google

Gemini 3.5 Flash

A fast Google model guide for cited-source search, creative content, coding help, and research workflows.

  • Cited search
  • Creative content
  • Coding help
Benchmarks
Design Arena · Code
Gemini 3.5 Flash
Elo
1,270
Preference win rate
54.3%
Tournament matches: 0

Reference benchmarks, not measurements of this deployment.

Open model guide
Image Generation

OpenAI

GPT Image 2

A new image model for high-fidelity visuals, UI mockups, multilingual text in images, and creative production tests.

  • Text-to-image
  • UI mockups
  • Creative visuals
Benchmarks

Not benchmarked

Open model guide
Search + Reasoning

xAI

Grok 4.5

A product-led xAI model guide for cited-source search, fact-checking, creative content, and research workflows.

  • Cited search
  • Fact-checking
  • Creative strategy
Benchmarks
Design Arena · Code
Grok 4.5
Elo
1,280
Preference win rate
51%
Tournament matches: 0

Reference benchmarks, not measurements of this deployment.

Open model guide
Long Context

Moonshot AI

Kimi K3

A strong model for long-context reasoning, creative workflows, coding tasks, and answers that need broad context.

  • Long documents
  • Coding workflows
  • Creative strategy
Benchmarks
Design Arena · Code
Kimi K3
Elo
1,366
Preference win rate
62.7%
Tournament matches: 0

Reference benchmarks, not measurements of this deployment.

Open model guide
Advanced Reasoning

Alibaba

Qwen3.8-Max

An Alibaba model guide for cited-source research, long-context analysis, coding tasks, and complex creative workflows.

  • Cited research
  • Coding workflows
  • Long-context analysis
Benchmarks

Not benchmarked

Open model guide
Search + Research

Z.ai

GLM 5.2

A Z.ai model guide for cited-source research, structured content, coding support, and complex professional prompts.

  • Cited research
  • Structured content
  • Coding support
Benchmarks

Benchmarks temporarily unavailable

Open model guide
Reasoning

DeepSeek

DeepSeek V4

A practical model for search, fact-checking, technical explanations, and structured content generation.

  • Search answers
  • Technical writing
  • Structured reasoning
Benchmarks

Benchmarks temporarily unavailable

Open model guide

Source: Design Arena (www.designarena.ai) via OpenRouter (openrouter.ai/rankings).

Source: Design Arena (www.designarena.ai) via OpenRouter (openrouter.ai/rankings).