deepseek-v4-flash speed results
deepseek-v4-flash real-world output speed and TTFT on TOKRACE, based on 56 anonymous runs.
Post to social channels, or use Markdown and badges for GitHub/README.
[](https://tokrace.com/en/model/deepseek-v4-flash)
How to read these metrics
Output tok/s: The clearest signal for long-form generation speed.
TTFT: Matters most for chatty or tool-heavy short requests.
Samples: More samples reduce one-off network and provider jitter.
· Data comes from voluntary anonymous sharing and contains speed metrics only · Medians reduce one-off jitter · Updates every 5 minutes
· Speed is affected by network, time of day and provider load · Methodology
FAQ
How fast is deepseek-v4-flash?
deepseek-v4-flash currently shows about 142 tok/s median output speed and 0.71s TTFT, based on 56 anonymous runs.
Should I treat this as a final benchmark?
Use it as directional evidence, not a single final benchmark. Prompt shape, network, time of day and provider load can all change the result.
Can I embed this result in an article or README?
Yes. This page provides Markdown and HTML badges. The badge image URL is https://tokrace.com/api/badge/model/deepseek-v4-flash?locale=en.