DeepSeek-V4-Flash-0731
The official release of DeepSeek-V4-Flash with enhanced agentic capabilities.
A specific language-model release identified in the supplied evidence as DeepSeek V4 Flash / Flash-0731, used for coding-agent and long-context workloads.
Pricing
Official site · Sep 6, 2026, 7:17 AM
Input / 1M
$0.44
Output / 1M
$1.32
Cached input / 1M
$0.014
Peak rate (01:00–04:00 and 06:00–10:00 UTC): $0.44/M cache-miss input, $1.32/M output, and $0.014/M cache-hit input. Off-peak rate: $0.22/M cache-miss input, $0.66/M output, and $0.007/M cache-hit input. Official page says these rates took effect August 16, 2026.
DeepSeek’s official API pricing table identifies the deepseek-v4-flash model version as DeepSeek-V4-Flash-0731. As of the stated effective date (August 16, 2026), it uses peak/off-peak token billing; normalized fields record peak prices.
Model Intelligence
Context window
1,000,000 tokens
Arena ranking
35
Benchmarkable
Yes
Model level
release
Intelligence Index
34.5
Coding Index
69.1
GPQA
0.91
HLE
0.39
SciCode
0.5
LCR
0.8
Recent stories
0 linked stories
No linked stories yet.