Deepseek V4 Flash
Observed throughTokscale Model Report
Deepseek V4 Flash model details and observed coding usage from Tokscale.
Reality Snapshot
Observed tokens
32.2B-17.2%
over 30 days
Estimated tracked spend
$647
Invoices may differ
Active developers
125-41
this month
30-day momentum
-17.2%
Compared with the prior 30-day window
Most-used client
OpenCode63%
of observed tokens
Observed Usage
Tokens observedDaily active developers
How developers use it
Usage by coding client
Sharevs 30 daysToken composition
Share of observed tokens
88%cache read
Cache read28.3B88%
Input3.8B12%
Output110.7M<1%
Reasoning28.2M<1%
Cache-heavy workloadMost observed tokens reuse cached context across requests.
Token Economics
Estimated spend profile
OpenCode$22935.33%
Pi$19029.41%
Hermes Agent$274.21%
CodeBuddy$14221.9%
Claude Code$416.39%
Gajae-Code$4.360.67%
Codex CLI$3.120.48%
Kilo$1.410.22%
Observed cost signals
Estimated spend per 1M tracked tokens$0.02
Cache-read share87.8%
Estimated cache savingsNot available
Typical workloadcontext-heavy
Estimated tracked spend calculated from Tokscale model and provider pricing data. Actual invoices may differ.
Model Momentum
-17.2%30-day growthRank: Observed model
Compared with the prior non-overlapping window
Fastest-growing client: OpenCode
Official Model Details
ProviderDeepseek
Tokscale Insight
Deepseek V4 Flash recorded 32.2B coding tokens across 125 developers during the latest 30-day Tokscale window.
Observed confidenceHigh






