# DeepSeek API vs. OpenRouter latency, measured daily

> Source: <https://latencyradar.com/compare/deepseek-vs-openrouter-latency/>
> Published: 2026-09-30 05:25:37+00:00

DEEPSEEK LATENCY BENCHMARK · Sept 2026

# DeepSeek direct or via OpenRouter?

We tested both from six cities. The faster option depends heavily on where you call from.

## What the data shows

1. Finding 1 ### Location matters more than the routeDirect was faster from Singapore and Mumbai. OpenRouter was often faster from Montreal. 
  - Singapore and Mumbai: direct 621–664 ms, routes 782–1,348 ms
  - Montreal: routes 422–692 ms, direct 732 ms
 [See the chart by city →](#by-city)
2. Finding 2 ### DeepSeek's direct API was more consistentDeepSeek's API varied less between locations: its slowest city was 36% above its fastest, against 89–219% through OpenRouter. 
  - City spread: 36% direct, 89–219% via OpenRouter
  - Slow days: 0 direct, 1 via OpenRouter
 [See the day-by-day chart →](#history)
3. Finding 3 ### Most of the wait isn't the networkConnection setup was only 11–19 ms. Most latency came after reaching the provider. 
  - Network: 11–19 ms
  - First token: 462–883 ms
 [See where the time goes →](#time-split)

## Time to first token, day by day

A dot marks a day when every compared city was well above that provider’s usual level.

## Show the numbers

| Day | DeepSeek direct | OpenRouter · Cloudflare | OpenRouter · Together | OpenRouter · DigitalOcean | 
|---|---|---|---|---|
| 27 August 2026 | — | — | 722 ms | 949 ms | 
| 28 August 2026 | — | 646 ms | 661 ms | 498 ms | 
| 29 August 2026 | — | — | 602 ms | 756 ms | 
| 31 August 2026 | — | — | 580 ms | 650 ms | 
| 1 September 2026 | — | — | 936 ms | 913 ms | 
| 2 September 2026 | — | — | 1586 ms | 540 ms | 
| 3 September 2026 | — | 796 ms | 784 ms | 827 ms | 
| 4 September 2026 | — | 654 ms | — | 772 ms | 
| 5 September 2026 | 714 ms | 533 ms | 765 ms | 639 ms | 
| 6 September 2026 | 493 ms | 599 ms | — | 666 ms | 
| 7 September 2026 | 617 ms | — | 875 ms | 799 ms | 
| 8 September 2026 | 538 ms | 626 ms | 1057 ms | 805 ms | 
| 9 September 2026 | 434 ms | 820 ms | 759 ms | 988 ms | 
| 10 September 2026 | 781 ms | 743 ms | 829 ms | 764 ms | 
| 11 September 2026 | 611 ms | 729 ms | 697 ms | 705 ms | 
| 12 September 2026 | 645 ms | 873 ms | 859 ms | 736 ms | 
| 13 September 2026 | 676 ms | 650 ms | 689 ms | 843 ms | 
| 14 September 2026 | 802 ms | 748 ms | 757 ms | 792 ms | 
| 15 September 2026 | 902 ms | 1274 ms | 687 ms | 752 ms | 
| 16 September 2026 | 733 ms | 714 ms | — | 695 ms | 
| 17 September 2026 | 594 ms | 745 ms | — | 861 ms | 
| 18 September 2026 | 732 ms | 727 ms | 5566 ms ▲ | 806 ms | 
| 19 September 2026 | 630 ms | 631 ms | 663 ms | 789 ms | 
| 20 September 2026 | 619 ms | 725 ms | — | 845 ms | 
| 21 September 2026 | 700 ms | 638 ms | 1138 ms | 722 ms | 
| 22 September 2026 | 638 ms | 637 ms | 814 ms | 881 ms | 

## Does your location matter?

DeepSeek direct was faster from Singapore and Mumbai in our tests.

- DeepSeek direct
- OpenRouter · Cloudflare
- OpenRouter · Together
- OpenRouter · DigitalOcean

## Show exact values

| City | DeepSeek direct | OpenRouter · Cloudflare | OpenRouter · Together | OpenRouter · DigitalOcean | 
|---|---|---|---|---|
| Amsterdam | 733 msslower 1.1 s | 537 msslower 1.2 s | 722 msslower 3.3 s | 608 msslower 931 ms | 
| San Francisco | 781 msslower 957 ms | 659 msslower 1.9 s | 627 msslower 5.8 s | 546 msslower 895 ms | 
| Montreal | 732 msslower 989 ms | 643 msslower 1.5 s | 692 msslower 5.5 s | 422 msslower 2.1 s | 
| Singapore | 621 msslower 866 ms | 782 msslower 1.8 s | 967 msslower 6.0 s | 939 msslower 2.1 s | 
| Tokyo | 575 msslower 1.2 s | 784 msslower 2.1 s | 724 msslower 5.9 s | 1081 msslower 2.0 s | 
| Mumbai | 664 msslower 883 ms | 1211 msslower 2.0 s | 1183 msslower 4.9 s | 1348 msslower 2.3 s | 

Bold is the lowest typical time in that city. "Slower" is the time 95% of the daily requests from that city finished within.

## Where does the wait happen?

How much of the wait a closer server could remove, and how much is the provider’s own queue and model.

## Direct vs OpenRouter

9–22 September 2026 · six test locations

| Metric | Direct | Cloudflare | Together | DigitalOcean | 
|---|---|---|---|---|
| Typical TTFT | 698 ms | 721 ms | 723 ms | 774 ms | 
| Location spread | 36% | 126% | 89% | 219% | 

Direct uses deepseek-v4-flash; the OpenRouter routes use deepseek-v4-flash-0731. [Methodology →](https://latencyradar.com/methodology/#ai-api-benchmarks)

## Show full benchmark details ↓

| Detail | DeepSeek direct | OpenRouter · Cloudflare | OpenRouter · Together | OpenRouter · DigitalOcean | 
|---|---|---|---|---|
| Model | deepseek-v4-flash | deepseek-v4-flash-0731 | deepseek-v4-flash-0731 | deepseek-v4-flash-0731 | 
| Slower requests, p95 (median city) | 973 ms | 1820 ms | 5649 ms | 2039 ms | 
| Fastest location | Tokyo · 575 ms | Amsterdam · 537 ms | San Francisco · 627 ms | Montreal · 422 ms | 
| Slowest location | San Francisco · 781 ms | Mumbai · 1211 ms | Mumbai · 1183 ms | Mumbai · 1348 ms | 
| Days when every city slowed together | 0 | 0 | 1 | 0 | 
| Requests in this window | 420 over 14 daily runs | 417 over 14 daily runs | 326 over 11 daily runs | 419 over 14 daily runs | 
| Latest controlled run | 7 September 2026 | 20 September 2026 | 13 September 2026 | 20 September 2026 | 

## How we test

We send the same streaming prompt from six cities to DeepSeek directly and through OpenRouter, then measure time to first visible token.

Same request on both sides: thinking off, temperature 0. The Fireworks and BaseTen routes are left out: one throttled us, the other never returned a visible token.

[Full testing methodology →](https://latencyradar.com/methodology/#ai-api-benchmarks)

## Good to know

- Network setup is the hop to OpenRouter's edge (11–19 ms from every city); the hop from OpenRouter to the provider's machines is counted as waiting for the model, which is why Mumbai still waits longer than Montreal.
- Direct sends deepseek-v4-flash, which DeepSeek has served with V4.1 Flash since 10 September (V4 Flash is retired), while the OpenRouter routes pin deepseek-v4-flash-0731, so the two sides have not been the same model since that day.
- Each OpenRouter request pins one provider with fallbacks off, and a city's run is discarded if any response came from another provider.
- The prompt is "Say 'ok' and nothing else." (max_tokens 256, temperature 0, thinking off), sent daily between 03:30 and 06:46 UTC, 5 requests per city after one discarded warm-up.
- We time the first visible token only; tokens per second, long prompts and thinking on are not measured.

## How does your API compare?

Test your endpoint from the same 6 cities used here and see your response time next to these numbers.

[Run a free latency test](https://latencyradar.com/#snapshot-form)

No account required · Takes about 30 seconds.