Claude 3 Haiku vs Gemini 1.5 Flash: Which AI Is Faster?

Two latency-optimized models compared on first-token wait and sustained throughput.

The short answer

There is no single winner — speed depends on the model's inference time and the network path from your connection to each provider. Run the live test below to measure Claude 3 Haiku and Gemini 1.5 Flash from your own location, on TTFT (first-token wait) and TPS (sustained throughput).

Run the live speed test →

What the numbers mean

TTFT (Time to First Token)
How long until the first token appears — the "reflex" you feel as lag before a reply starts. Lower is better.
TPS (Tokens Per Second)
How fast text streams once it starts. Higher is better. See the full methodology.

Other comparisons