# LLM Latency Benchmark Report

**Generated**: 2026-01-27 05:09:00

## Summary

| Model | TTFT (median) | Total (median) | Tokens/sec | Success |
|-------|---------------|----------------|------------|---------|
| anthropic/claude-haiku-4-5-20251001 | 569ms | 3851ms | 91.5 | 100% |
| xai/grok-4-1-fast-non-reasoning | 712ms | 3088ms | 127.7 | 100% |
| gemini/gemini-2.5-flash-lite | 2323ms | 3790ms | 214.7 | 100% |

## Detailed Results

### anthropic/claude-haiku-4-5-20251001

**TTFT (Time to First Token)**
- Min: 548ms
- Max: 591ms
- Mean: 569ms
- Median: 569ms
- Stdev: 30ms

**Total Response Time**
- Min: 3755ms
- Max: 3946ms
- Mean: 3851ms
- Median: 3851ms
- Stdev: 135ms

**Individual Runs**

- Run 1: TTFT=548ms, Total=3946ms, Tokens=300, 88.3 tok/s
- Run 2: TTFT=591ms, Total=3755ms, Tokens=300, 94.8 tok/s

### xai/grok-4-1-fast-non-reasoning

**TTFT (Time to First Token)**
- Min: 328ms
- Max: 1097ms
- Mean: 712ms
- Median: 712ms
- Stdev: 544ms

**Total Response Time**
- Min: 2640ms
- Max: 3536ms
- Mean: 3088ms
- Median: 3088ms
- Stdev: 633ms

**Individual Runs**

- Run 1: TTFT=1097ms, Total=3536ms, Tokens=301, 123.4 tok/s
- Run 2: TTFT=328ms, Total=2640ms, Tokens=305, 131.9 tok/s

### gemini/gemini-2.5-flash-lite

**TTFT (Time to First Token)**
- Min: 1261ms
- Max: 3384ms
- Mean: 2323ms
- Median: 2323ms
- Stdev: 1501ms

**Total Response Time**
- Min: 2415ms
- Max: 5165ms
- Mean: 3790ms
- Median: 3790ms
- Stdev: 1945ms

**Individual Runs**

- Run 1: TTFT=3384ms, Total=5165ms, Tokens=300, 168.4 tok/s
- Run 2: TTFT=1261ms, Total=2415ms, Tokens=301, 260.9 tok/s

## Configuration

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