# LLM Latency Benchmark Report

**Generated**: 2026-01-27 05:46:35

## Summary

| Model | TTFT (median) | Total (median) | Tokens/sec | Success |
|-------|---------------|----------------|------------|---------|
| anthropic/claude-sonnet-4-5-20250929 | 1171ms | 8628ms | 40.6 | 100% |
| anthropic/claude-opus-4-5-20251101 | 1592ms | 7562ms | 50.4 | 100% |
| gemini/gemini-3-pro-preview | 101880ms | 103329ms | 191.6 | 100% |

## Detailed Results

### anthropic/claude-sonnet-4-5-20250929

**TTFT (Time to First Token)**
- Min: 994ms
- Max: 1628ms
- Mean: 1246ms
- Median: 1171ms
- Stdev: 249ms

**Total Response Time**
- Min: 8477ms
- Max: 9052ms
- Mean: 8689ms
- Median: 8628ms
- Stdev: 238ms

**Individual Runs**

- Run 1: TTFT=1628ms, Total=8791ms, Tokens=300, 41.9 tok/s
- Run 2: TTFT=1093ms, Total=8477ms, Tokens=300, 40.6 tok/s
- Run 3: TTFT=994ms, Total=8628ms, Tokens=300, 39.3 tok/s
- Run 4: TTFT=1171ms, Total=8499ms, Tokens=300, 40.9 tok/s
- Run 5: TTFT=1342ms, Total=9052ms, Tokens=300, 38.9 tok/s

### anthropic/claude-opus-4-5-20251101

**TTFT (Time to First Token)**
- Min: 1491ms
- Max: 1689ms
- Mean: 1590ms
- Median: 1592ms
- Stdev: 80ms

**Total Response Time**
- Min: 7037ms
- Max: 8331ms
- Mean: 7591ms
- Median: 7562ms
- Stdev: 475ms

**Individual Runs**

- Run 1: TTFT=1533ms, Total=8331ms, Tokens=300, 44.1 tok/s
- Run 2: TTFT=1491ms, Total=7562ms, Tokens=300, 49.4 tok/s
- Run 3: TTFT=1645ms, Total=7037ms, Tokens=300, 55.6 tok/s
- Run 4: TTFT=1592ms, Total=7382ms, Tokens=300, 51.8 tok/s
- Run 5: TTFT=1689ms, Total=7642ms, Tokens=300, 50.4 tok/s

### gemini/gemini-3-pro-preview

**TTFT (Time to First Token)**
- Min: 4651ms
- Max: 138891ms
- Mean: 81646ms
- Median: 101880ms
- Stdev: 52616ms

**Total Response Time**
- Min: 6163ms
- Max: 140436ms
- Mean: 83299ms
- Median: 103329ms
- Stdev: 52633ms

**Individual Runs**

- Run 1: TTFT=4651ms, Total=6163ms, Tokens=296, 195.8 tok/s
- Run 2: TTFT=54397ms, Total=56218ms, Tokens=296, 162.5 tok/s
- Run 3: TTFT=101880ms, Total=103329ms, Tokens=296, 204.2 tok/s
- Run 4: TTFT=138891ms, Total=140436ms, Tokens=296, 191.6 tok/s
- Run 5: TTFT=108413ms, Total=110350ms, Tokens=296, 152.8 tok/s

## Configuration

- **Prompt**: "Explain how a CPU cache works in 3 paragraphs."
- **Max Tokens**: 300
- **Timeout**: 60s
