# Percentile (p50, p99)

> Percentile (p50, p99): pN is the value N% of measurements fall at or below: p50 is the median, p99 is beaten by all but the slowest 1%. Used for latency…

Canonical: https://llmdojo.dev/glossary/percentile

Also called percentiles, percentile, p99, p50.

pN is the value N% of measurements fall at or below: p50 is the median, p99 is beaten by all but the slowest 1%. Used for latency because the mean hides the tail.

Learn it in the lesson: [Latency is a distribution](https://llmdojo.dev/learn/inference-optimization/measuring-inference/latency-is-a-distribution) (Inference and Model Optimization · Measuring Inference: Latency, Throughput and Goodput).

## Interview questions about Percentile (p50, p99)

- [Why report p50 and p99 latency rather than the mean?](https://llmdojo.dev/interview/q/why-report-p50-and-p99-latency-rather-than-the-mean)
- [A provider's status page shows a p50 TTFT of 180 ms. You need 300 ms for 9 in 10 of your users. What do you do before signing up?](https://llmdojo.dev/interview/q/a-provider-s-status-page-shows-a-p50-ttft-of-180-ms)
- [Your AI chatbot takes too long to respond. What do you do?](https://llmdojo.dev/interview/q/your-ai-chatbot-takes-too-long-to-respond-what-do-you-do)
- [Implement FP32-to-int8 quantization in NumPy with a scale, zero point and clipping, and the matching dequantizer.](https://llmdojo.dev/interview/q/implement-fp32-to-int8-quantization-in-numpy-with-a-scale-zero-point)
- [What does a fair inference benchmark hold fixed?](https://llmdojo.dev/interview/q/what-does-a-fair-inference-benchmark-hold-fixed)
- [Design optimised inference pipelines, and tell me how you would benchmark them.](https://llmdojo.dev/interview/q/design-optimised-inference-pipelines-and-tell-me-how-you-would-benchmark-them)
- [Your answers must start within one second. How do you budget it?](https://llmdojo.dev/interview/q/your-answers-must-start-within-one-second-how-do-you-budget-it)
- [In the recall stage of a recommender, design a model that chooses how many candidates K to retrieve per request given latency and compute cost.](https://llmdojo.dev/interview/q/in-the-recall-stage-of-a-recommender-design-a-model-that-chooses)

## Related
- [Time to first token (TTFT)](https://llmdojo.dev/glossary/ttft)
- [Time per output token (TPOT)](https://llmdojo.dev/glossary/tpot)
- [Inter-token latency](https://llmdojo.dev/glossary/itl)
- [Throughput](https://llmdojo.dev/glossary/throughput)
- [Goodput](https://llmdojo.dev/glossary/goodput)
- [Service level objective (SLO)](https://llmdojo.dev/glossary/slo)
- [Time to publish](https://llmdojo.dev/glossary/time-to-publish)
- [Glossary of every term](https://llmdojo.dev/glossary)