AI model comparison
Pick 2 to 3 models and compare them across every independent Artificial Analysis and Design Arena test. The best value in each row is highlighted.
Models to compare
Ready-made comparisons
Claude Opus 4.6 vs Deepseek V4 ProClaude Opus 4.6 vs Gemini 3.6 FlashClaude Opus 4.6 vs GPT 5.5Claude Opus 4.6 vs Kimi K3Claude Opus 4.6 vs Qwen 3.8 MaxDeepseek V4 Pro vs GLM 5.3 FlashXDeepseek V4 Pro vs GPT 5.5Deepseek V4 Pro vs Qwen 3.8 MaxGemini 3.6 Flash vs GPT 5.5Gemini 3.6 Flash vs Qwen 3.8 MaxGPT 5.5 vs Grok 4.5GPT 5.5 vs Kimi K3
Specifications of all text models
| Model | API ID | Vendor | Context | Max output | Images | Multiplier |
|---|---|---|---|---|---|---|
| Claude Fable 5 | claude-fable-5 | Anthropic | 1M tokens | 128K tokens | yes | 10× – 14× |
| Claude Fable 5.1 | claude-fable-5.1 | Anthropic | 1M tokens | 128K tokens | yes | 13× – 14× |
| Claude Haiku 4.5 | claude-haiku-4-5 | Anthropic | 200K tokens | 8.2K tokens | yes | 1.6× |
| Claude Opus 4.6 | claude-opus-4-6 | Anthropic | 1M tokens | 128K tokens | yes | 1.8× |
| Claude Opus 4.7 | claude-opus-4-7 | Anthropic | 1M tokens | 128K tokens | yes | 1.8× |
| Claude Opus 4.8 | claude-opus-4-8 | Anthropic | 1M tokens | 128K tokens | yes | 1.8× |
| Claude Opus 5 | claude-opus-5 | Anthropic | 1M tokens | 128K tokens | yes | 1.9× |
| Claude Sonnet 4.6 | claude-sonnet-4-6 | Anthropic | 1M tokens | 128K tokens | yes | 1.7× |
| Claude Sonnet 5 | claude-sonnet-5 | Anthropic | 1M tokens | 128K tokens | yes | 1.7× |
| Deepseek V4 Pro | deepseek-v4-pro | DeepSeek | 1.05M tokens | 393K tokens | no | 1.8× – 2× |
| Deepseek V4.1 Flash | deepseek-v4.1-flash | DeepSeek | 1.05M tokens | 393.2K tokens | yes | 1.7× |
| Free Claude Opus 5 | free-claude-opus-5 | Anthropic | 1M tokens | 128K tokens | yes | 1.9× |
| Free Gemini 3.8 Flash | free-gemini-3.8-flash | Gemini | 1M tokens | 65.5K tokens | yes | 1.7× |
| Free GLM 5.3 Flash | free-glm-5.3-flash | Z-AI | 1M tokens | 131.1K tokens | yes | 1.9× |
| Free GPT-6 Astra | free-gpt-6-astra | OpenAI | 1M tokens | 128K tokens | yes | 6× |
| Gemini 3.1 Flash Lite | gemini-3.1-flash-lite | Gemini | 1M tokens | 65.5K tokens | yes | 1.5× |
| Gemini 3.1 Pro Preview | gemini-3.1-pro-preview | Gemini | 1M tokens | 65.5K tokens | yes | 1.7× – 1.8× |
| Gemini 3.6 Flash | gemini-3.6-flash | Gemini | 1.05M tokens | 65.5K tokens | yes | 1.4× – 1.7× |
| Gemini 3.7 Flash | gemini-3.7-flash | Gemini | 1M tokens | 65.5K tokens | yes | 1.4× – 1.7× |
| Gemini 3.8 Flash | gemini-3.8-flash | Gemini | 1M tokens | 65.5K tokens | yes | 1.4× – 1.7× |
| Gemini Embedding 001 | gemini-embedding-001 | OpenAI | 20K tokens | 20K tokens | no | 1.35× |
| Gemini Embedding 2 | gemini-embedding-2 | Gemini | 8.2K tokens | 8.2K tokens | no | 1.4× |
| GLM 5.2 | glm-5.2 | Z-AI | 1M tokens | 128K tokens | no | 2× – 3× |
| GLM 5.3 | glm-5.3 | Z-AI | 1M tokens | 128K tokens | no | 2× – 4× |
| GLM 5.3 Flash | glm-5.3-flash | Z-AI | 1M tokens | 131.1K tokens | yes | 1.5× – 1.7× |
| GLM 5.3 FlashX | glm-5.3-flashx | Z-AI | 1.05M tokens | 131.1K tokens | no | 1.6× |
| GPT 5.5 | gpt-5.5 | OpenAI | 1.05M tokens | 128K tokens | yes | 1.8× |
| GPT-4o mini | gpt-4o-mini | OpenAI | 128K tokens | 16.4K tokens | yes | 1.5× |
| GPT-5.6 Luna | gpt-5.6-luna | OpenAI | 1M tokens | 128K tokens | yes | 1.6× |
| GPT-5.6 Sol | gpt-5.6-sol | OpenAI | 1M tokens | 128K tokens | yes | 1.8× – 1.9× |
| GPT-5.6 Terra | gpt-5.6-terra | OpenAI | 1M tokens | 128K tokens | yes | 1.7× – 1.8× |
| GPT-6 Astra | gpt-6-astra | OpenAI | 1M tokens | 128K tokens | yes | 5× – 6× |
| Grok 4.5 | grok-4.5 | xAI | 500K tokens | 128K tokens | yes | 1.7× |
| Grok 4.6 | grok-4.6 | xAI | 500K tokens | 128K tokens | yes | 1.7× |
| Grok Build 0.1 | grok-build-0.1 | xAI | 256K tokens | 64K tokens | yes | 1.6× |
| Grok Imagine Video 1.5 | grok-imagine-video-1.5 | xAI | — | — | yes | 8× |
| H3 | hailuo-3 | MiniMax | — | — | yes | 1× |
| Hy-MT2-1.8B | hy-mt2-1.8b | Tencent | 8.2K tokens | 4.1K tokens | no | 1.4× |
| Hy-MT2-30B-A3B | hy-mt2-30b-a3b | Tencent | 8.2K tokens | 4.1K tokens | no | 1.8× |
| Hy-MT2-7B | hy-mt2-7b | Tencent | 8.2K tokens | 4.1K tokens | no | 1.8× |
| Hy3 | hy3 | Tencent | 262.1K tokens | 262.1K tokens | no | 1.7× – 2.1× |
| Hy4 preview | hy4-preview | Tencent | 1M tokens | 64K tokens | no | 1.5× |
| Kimi K2.7 Code | kimi-k2-7-code | MoonshotAI | 262.1K tokens | 16.4K tokens | yes | 3× – 4× |
| Kimi K3 | kimi-k3 | MoonshotAI | 1M tokens | 128K tokens | yes | 6× – 10× |
| LongCat 2.0 | longcat-2.0 | Meituan | 1M tokens | 262.1K tokens | no | 4× |
| Mercury 2.5 | mercury-2.5 | Inception | 260K tokens | 65.5K tokens | no | 1.3× |
| MiMo-V2.5 | mimo-v2.5 | Xiaomi | 1M tokens | 128K tokens | yes | 1.6× |
| MiMo-V2.5-Pro | mimo-v2.5-pro | Xiaomi | 1M tokens | 128K tokens | no | 1.7× |
| MiniMax M3 | minimax-m3 | MiniMax | 1M tokens | 16K tokens | yes | 1.5× |
| Muse Spark 1.2 | muse-spark-1.2 | Meta | 1.05M tokens | 128K tokens | yes | 9× |
| Muse Spark 1.3 | muse-spark-1.3 | Meta | 1M tokens | 128K tokens | yes | 9× |
| Omni | omni | Gemini | — | — | yes | 1× |
| Pareto | pareto | Unbiased | 262.1K tokens | 131.1K tokens | no | 10× |
| Qwen 3.7 Max | qwen-3.7-max | Qwen | 1M tokens | 131.1K tokens | no | 2.5× – 3× |
| Qwen 3.7 Plus | qwen-3.7-plus | Qwen | 1M tokens | 65.5K tokens | yes | 2× – 4× |
| Qwen 3.8 Flash | qwen-3.8-flash | Qwen | 1M tokens | 131.1K tokens | yes | 1.4× – 2× |
| Qwen 3.8 Max | qwen-3-8-max | Qwen | 1M tokens | 65.5K tokens | yes | 7× – 10× |
| Seedance 2.0 | seedance-2.0 | ByteDance | — | — | yes | 3× |
| Seedance 2.0 Fast | seedance-2.0-fast | ByteDance | — | — | yes | 1× |
| Seedance 2.0 Mini | seedance-2.0-mini | ByteDance | — | — | yes | 1× |
| Solar Pro 4 | solar-pro-4 | Upstage | 524.3K tokens | 131.1K tokens | no | 1.4× |
| Ternary Bonsai 2 27B | ternary-bonsai-2-27b | PrismML | 262.1K tokens | 32.8K tokens | yes | 1.5× |
| Text Embedding 3 Large | text-embedding-3-large | OpenAI | 8.2K tokens | 8.2K tokens | no | 1.2× |
| Text Embedding 3 Small | text-embedding-3-small | OpenAI | 8.2K tokens | 8.2K tokens | no | 1× |
| Composer 2.5 | composer-2.5 | Cursor | 1M tokens | 128K tokens | yes | 1× |
| HappyHorse 1.1 | happyhorse-1.1 | Alibaba | — | — | yes | 1× |
| Seedance 2.5 | seedance-2.5 | ByteDance | — | — | yes | 1× |
Values come from the Tokenator routing configuration. Per-model details live on each model's page in the catalog; token counting is covered in the documentation.