Model Comparison
Four views of the current index — capability, context, speed, and cost. Use the tabs to navigate. Filter by provider using the selector above.
Which models give you the most intelligence for your money?
Plotting benchmark capability against price reveals which models deliver frontier intelligence at frontier prices, and which deliver strong capability at a fraction of the cost. Models in the bottom-right are expensive for their capability. Models in the top-left are the value plays.
High value zone
Anthropic
OpenAI
MiniMax
DeepSeek
Mistral
Google
Meta
Z.ai
xAI
Qwen
MoonshotAI
Scores reflect the Artificial Analysis Intelligence Index v5, launched 27 October 2026. Earlier versions are not comparable.
Not yet scored on v5: Mistral Medium 3.5, GPT-5.4 Mini, Qwen 3.7 Plus, GLM-5.3-Flash.
25 models
| Model | Blended |
|---|---|
| Claude Haiku 5.5 | $0.220 |
| GPT-6 Luna | $0.220 |
| GLM-5.3-Flash | $0.255 |
| MiniMax M3.1 Flash | $0.285 |
| DeepSeek V4.1 Flash | $0.285 |
| MiniMax M3 | $0.570 |
| Qwen 3.7 Plus | $0.760 |
| DeepSeek V4-Pro | $1.056 |
| Mistral Large 4 | $1.120 |
| Gemini 3.8 Flash | $1.650 |
| GPT-5.4 Mini | $1.875 |
| Muse Spark 1.3 | $2.150 |
| GPT-6.1 Terra | $2.200 |
| GLM-5.3 | $2.300 |
| Grok 4.7 | $3.200 |
| Qwen 3.8 Max | $3.200 |
| Mistral Medium 3.5 | $3.300 |
| Kimi K3.1 Pro | $4.400 |
| Gemini 4 Argon | $4.400 |
| GPT-6.1 Sol | $4.400 |
| Claude Sonnet 5.5 | $4.400 |
| Muse Wonder | $8.800 |
| Claude Opus 5.5 | $8.800 |
| GPT-6 Astra | $22.000 |
| Claude Fable 5.1 | $22.000 |