Latency & Real-Time
Reference infra, us-east
Reference infra, us-east
| # | Details | ||||
|---|---|---|---|---|---|
| 1 | View Gemma 4 26B A4BGemma 4 26B A4B | 86 ms | 8 / 8 | Real-time | |
| 2 | View Holo3 35B A3BHolo3 35B A3B | 92 ms | 8 / 8 | Real-time | |
| 3 | View Gemini 3 FlashGemini 3 Flash | 139 ms | 4 / 5 | Real-time | |
| 4 | View Qwen3.6 27BQwen3.6 27B | 140 ms | 7 / 8 | Real-time | |
| 5 | View Gemma 4 31BGemma 4 31B | 141 ms | 8 / 8 | Real-time | |
| 6 | View Claude Haiku 4.5Claude Haiku 4.5 | 141 ms | 3 / 3 | Real-time | |
| 7 | View GPT-5.4 nanoGPT-5.4 nano | 1.4 s | 0 / 4 | Batch | |
| 8 | View Qwen3.6 35B A3BQwen3.6 35B A3B | 2.4 s | 0 / 8 | Batch | |
| 9 | View GPT-5.4 miniGPT-5.4 mini | 2.6 s | 0 / 4 | Batch | |
| 10 | View Claude Sonnet 4.6Claude Sonnet 4.6 | 7.5 s | 0 / 3 | Batch | |
| 11 | View Gemini 3.1 ProGemini 3.1 Pro | 7.6 s | 0 / 5 | Batch | |
| 12 | View GPT-5.4GPT-5.4 | 8.9 s | 0 / 4 | Batch | |
| 13 | View Claude Opus 4.6Claude Opus 4.6 | 11 s | 0 / 3 | Batch |