LLM Benchmark
Compare models across speed, cost, quality, and reliability
Configure Benchmark
Select Models
OpenAI
Anthropic
Gemini
Settings
Request Count100
Prompt TypeMixed (summarization + Q&A + code)
Timeout (ms)5000
Latency Comparison (ms)
Performance Radar
Full Results
| Model | Success Rate | Avg Latency | P95 | Cost/1K tokens | Throughput (req/s) |
|---|---|---|---|---|---|
🥇gpt-4o-mini | 98.2% | 680ms | 1200ms | $0.15 | 45 |
🥈claude-3-haiku | 97.8% | 720ms | 1350ms | $0.25 | 38 |
🥉gemini-flash | 96.5% | 450ms | 890ms | $0.075 | 62 |