AI Model Intelligence
Independent benchmarks and enterprise AI analysis across 693+ models from 96 providers — helping technology leaders, software engineers, and data science teams select the right AI platform for their business.
Total Models
Language Models
Media Models
Avg Intelligence
Providers
Intelligence Index
Composite intelligence score across all major benchmarks. Toggle between all models, open-weights only, or proprietary only.
Intelligence vs Cost
Plots Intelligence Index against price per 1M tokens. Upper-left models offer the best value.
Intelligence vs Cost — Zoomed In
Same chart with the top 10% most expensive outliers removed.
Image & Video Leaderboard
ELO ratings from head-to-head comparisons of image and video generation models.
| Model | ELO | 95% CI | Appearances |
|---|---|---|---|
| BGenFlare 2.0 | 1,329 | -8/8 | 7,213 |
| MHailuo 02 0616 | 1,287 | -14/14 | 2,353 |
| BGenFlare | 1,276 | -11/11 | 3,557 |
| TTeleVideo 2.0 | 1,271 | -10/10 | 4,680 |
| VAvenger 0.5 Pro | 1,271 | -10/10 | 4,200 |
 GPT Image 1.5 (high) | 1,268 | -12/12 | 3,933 |
| MHailuo 2.3 Fast | 1,253 | -10/10 | 4,653 |
| SRiverflow 2.0 | 1,252 | -12/12 | 4,236 |
| VAvenger 0.5 | 1,245 | -14/14 | 2,449 |
| VVidu Q2 Turbo | 1,241 | -11/11 | 3,573 |
| KKling 3.0 Pro | 1,240 | -11/11 | 3,681 |
| VVidu Q2 Pro | 1,237 | -11/11 | 3,677 |
| KKling 3.0 Omni Pro | 1,234 | -12/12 | 3,402 |
| PPixVerse V5.6 | 1,233 | -12/12 | 3,271 |
| Xgrok-imagine-video | 1,227 | -10/10 | 5,279 |
 Runway Gen-4.5 | 1,225 | -8/8 | 8,057 |
| VVidu Q3 Pro | 1,225 | -9/9 | 5,298 |
 Veo 3.1 Preview | 1,221 | -11/11 | 3,865 |
| Veo 3 | 1,221 | -12/12 | 3,121 |
| Nano Banana Pro (Gemini 3 Pro Image) | 1,220 | -13/13 | 2,990 |
Frontier Intelligence Over Time
Tracks the highest-scoring model each month by Intelligence Index.
Intelligence Evaluations
The hardest subset of GPQA, filtered for questions where experts agree and non-experts struggle.
Pricing: Input vs Output
View AllCompares input and output token pricing (per 1M tokens) for the most affordable models.
Cost Efficiency
Intelligence Index plotted against cost per 1M tokens.
Key Findings
Category Overview
Latest Updates
Text-to-Image
Compare AI image generation models used in marketing, e-commerce, and creative design workflows.
Performance
Analyze speed, latency, and throughput metrics critical for real-time enterprise applications.
Coding
Compare models on code generation, debugging, and software engineering tasks.