Live Rankings
AI Model Leaderboard
Comprehensive rankings of AI models based on real-world performance data from the Zip platform. Search across all 168 models.
Top 10 Charts
Intelligence Index
Composite score from real agentic tasks
Speed
Output tokens per second
Price (Blended)
USD per 1M tokens • 7:2:1 ratio
Full Rankings
Showing 151-168 of 168 models
Click column headers to sort
| # | Model | Creator | Intelligence | Speed | Price | Popular | Type | Compare |
|---|---|---|---|---|---|---|---|---|
| 151 | Qwen3.6 27B Qwen (Alibaba) | 0.0 | 0.0 tok/s | $0.92 | 4.0 | Proprietary | | |
| 152 | Qwen3.6 35B A3B Qwen (Alibaba) | 0.0 | 0.0 tok/s | $0.40 | 4.0 | Proprietary | | |
| 153 | Qwen3.6 Flash Qwen (Alibaba) | 0.0 | 0.0 tok/s | $0.47 | 4.5 | Proprietary | | |
| 154 | Qwen3.6 Plus Qwen (Alibaba) | 0.0 | 0.0 tok/s | $0.81 | 4.5 | Proprietary | | |
| 155 | Qwen3.7 Max Qwen (Alibaba) | 0.0 | 0.0 tok/s | $2.00 | 4.5 | Proprietary | | |
| 156 | Qwen3.7 Plus Qwen (Alibaba) | 0.0 | 0.0 tok/s | $0.76 | 4.5 | Proprietary | | |
| 157 | R1 DeepSeek | 82.5 | 38.7 tok/s | $1.24 | 6.5 | Proprietary | | |
| 158 | R1 0528 DeepSeek | 0.0 | 0.0 tok/s | $0.99 | 6.5 | Proprietary | | |
| 159 | R1 Distill Llama 70B DeepSeek | 0.0 | 0.0 tok/s | $0.73 | 6.5 | Proprietary | | |
| 160 | R1 Distill Qwen 32B DeepSeek | 0.0 | 0.0 tok/s | $0.29 | 6.5 | Proprietary | | |
| 161 | Sonar Pro Perplexity | 0.0 | 0.0 tok/s | $6.60 | 4.0 | Proprietary | | |
| 162 | Sonar Pro Search Perplexity | 0.0 | 0.0 tok/s | $6.60 | 4.0 | Proprietary | | |
| 163 | Z.ai: GLM 4.6 Z.ai | 0.0 | 0.0 tok/s | $0.82 | 4.0 | Proprietary | | |
| 164 | Z.ai: GLM 4.7 Z.ai | 0.0 | 0.0 tok/s | $0.81 | 4.0 | Proprietary | | |
| 165 | Z.ai: GLM 4.7 Flash Z.ai | 0.0 | 0.0 tok/s | $0.16 | 4.0 | Proprietary | | |
| 166 | Z.ai: GLM 5 Z.ai | 0.0 | 0.0 tok/s | $1.00 | 4.0 | Proprietary | | |
| 167 | Z.ai: GLM 5 Turbo Z.ai | 0.0 | 0.0 tok/s | $2.04 | 4.0 | Proprietary | | |
| 168 | Z.ai: GLM 5.1 Z.ai | 0.0 | 0.0 tok/s | $1.61 | 4.0 | Proprietary | |
How to interpret these rankings
Intelligence
Composite score from real agentic tasks including coding, research, analysis, and planning. Higher is better.
Speed
Output tokens per second using OpenAI tokens as standard measurement. Higher is better.
Price
Blended price per 1M tokens assuming 7:2:1 ratio of cache hit, input, and output. Lower is better.