DeepSeek V4 Flash Fast

deepseek/deepseek-v4-flash-fast

low-latency V4 Flash deployment.

cmd --model deepseek/deepseek-v4-flash-fast
Intelligence index
not yet scored
Output speed
not yet scored
Input
$0.28 /M
Output
$0.56 /M
Cache read
$0.07 /M
Agent-loop cost
$0.13 /M in
Context window
1M tokens
Released
Modalities

vs. the lineup

DeepSeek V4 Flash Fast beside its stablemates and nearest rivals. The ◆ marks the best value in each column across every row shown.

pin a rival:
ModelIntelligenceCodingSpeedInput $/MOutput $/MBlended $/MContext
GLM-5.3 Flash57.571.541.8$0.15$0.50$0.241.05M
DeepSeek V4 Pro (latest)53.268.861$0.66$1.98$0.991M
DeepSeek V4 Flash (latest)5269.1129$0.22$0.66$0.331M
Gemini 3.5 Flash Lite37.449.3368.8$0.30$2.50$0.851M
DeepSeek V4 Flash Fast$0.28$0.56$0.351M
DeepSeek V4 Flash Vision (exp)$0.22$0.66$0.331M
Kimi K3pinned59.776.238.4$3$15$61M
Kimi K2.7 Codepinned4360.839.5$0.95$4$1.71256K

usage calculator

How far a month of credits goes on DeepSeek V4 Flash Fast.

Input tokensfresh prompt
800
Output tokensmodel reply
180
Cache read tokensre-read context
50K
cost / request $0.0038 · in $0.28 · out $0.56 · cache $0.07 per M
fresh input 6%output 3%cache reads 92%
Requests / 30 days
2.6K
$10 credits ÷ $0.0038 per request
~523 quick fixes~105 bug fixes~17 feature PRs

what real work costs

Real coding tasks priced end to end on DeepSeek V4 Flash Fast, from a quick lookup to a full-repo agent run.

One agent task
$0.03
180K in at 75% cache hit, 12K out
What you pay
$0.15 /M
all-in across every token that task touched
Sticker input
$0.28 /M
cache reads bill at $0.07 /M instead
TaskTokens in · outDeepSeek V4 Flash FastGLM-5.3 FlashClaude Haiku 4.5
Quick lookup / one-liner8K · 1K$0.0013$0.0008$0.0065
Review a 500-line PR60K · 4K$0.01$0.0060$0.04
Fix a bug (agent loop)180K · 12K$0.03$0.02$0.12
Refactor a module320K · 20K$0.05$0.03$0.21
Full-repo agent run900K · 45K$0.13$0.07$0.49

frequently asked

How much does DeepSeek V4 Flash Fast cost?
$0.28/M input and $0.56/M output, cache reads $0.07/M. In an agent loop most input is cache-read, so the effective input rate is about $0.13/M.
Which plan do I need?
Available on Go and above.
How do I switch to it?
Run cmd --model deepseek/deepseek-v4-flash-fast, or type /model in a session and pick it. You can switch mid-session without losing context.

Ship code that matches your taste

Command Code is the AI coding agent that continuously learns your taste. Start for $1.

Benchmarks from Artificial Analysis (v4.1)commandcode.ai/models/deepseek-v4-flash-fast