---
id: provider-cache-hit-rate/2026-07
series: provider-cache-hit-rate
title: Prompt cache hit rate by provider, July 2026
period: 2026-07
period_start: 2026-07-01
period_end: 2026-07-31
updated: 2026-11-11
license: CC BY 4.0
canonical: https://www.requesty.ai/data/provider-cache-hit-rate/2026-07
latest: https://www.requesty.ai/data/provider-cache-hit-rate
---

# Prompt cache hit rate by provider, July 2026

> Sference served 92.6% of input tokens from cache in July 2026, the highest among major providers. The median major provider served 72.3%.

Question: What share of input tokens do LLM providers serve from prompt cache?

## Key findings

1. Sference leads at 92.6%; Mistral is lowest at 13.7%.
2. 4 of 16 major providers served at least 80% of input tokens from cache.
3. xAI moved the most since June 2026: cache hit rate went from 36.9% to 74.2% (+37.3 pp).

## Data

| Provider | Cache hit rate | Traffic share | Organizations (band) |
| --- | --- | --- | --- |
| Sference | 92.62% | 1.70% | 100-499 |
| DeepSeek | 88.45% | 9.30% | 500+ |
| Fireworks | 86.63% | 3.57% | 500+ |
| Anthropic | 84.49% | 17.69% | 500+ |
| Vertex (Claude) | 77.08% | 3.53% | 500+ |
| TensorX | 76.62% | 2.45% | 500+ |
| xAI | 74.20% | 1.01% | 100-499 |
| Azure | 72.95% | 7.01% | 500+ |
| Bedrock | 71.70% | 3.40% | 500+ |
| Nebius | 70.14% | 1.18% | 500+ |
| DeepInfra | 55.16% | 1.65% | 500+ |
| Novita | 38.68% | 1.49% | 500+ |
| Alibaba | 35.42% | 0.62% | 100-499 |
| OpenAI | 29.68% | 20.67% | 500+ |
| Vertex (Gemini) | 27.18% | 22.02% | 500+ |
| Mistral | 13.66% | 0.85% | 500+ |

## Method

- Cache hit rate is cached input tokens divided by input tokens, the same definition /rankings uses.
- Providers that report no cached tokens are left out of the chart. Only providers with at least 0.5% of published traffic are shown.
- A row is published only when it has at least 20 requests, at least 5 distinct organizations, and no single organization above 80% of its traffic. These are the same floors /rankings uses. Rows below the floors are withheld, not estimated.
- Only shares, rates, medians and organization bands are published. Absolute request, token and spend totals stay private.

## Caveats

- Hit rate depends on workload. Long agent sessions with a stable prefix cache far better than one-off requests.
- Providers report cached tokens differently. Some only count explicit cache reads.

## Cite as

Requesty (2026). Prompt cache hit rate by provider, July 2026. Requesty Data. https://www.requesty.ai/data/provider-cache-hit-rate/2026-07

```bibtex
@misc{requesty_provider_cache_hit_rate_2026_07,
  author       = {{Requesty}},
  title        = {Prompt cache hit rate by provider, July 2026},
  year         = {2026},
  howpublished = {\url{https://www.requesty.ai/data/provider-cache-hit-rate/2026-07}},
  note         = {Requesty Data}
}
```

Source: Requesty production gateway. License: CC BY 4.0 (https://creativecommons.org/licenses/by/4.0/).
Downloads: [JSON](https://www.requesty.ai/data/provider-cache-hit-rate/2026-07/data.json) · [CSV](https://www.requesty.ai/data/provider-cache-hit-rate/2026-07/data.csv) · [Markdown](https://www.requesty.ai/data/provider-cache-hit-rate/2026-07/data.md)