modelbenchmark.io

DeepSeek V3.2 Exp Thinking

DeepSeek · deepseek-v3-2-exp-thinking

Compare

DeepSeek chat model for instruction following, coding, and analysis

Specification

most-agreed values

Context
164K
Max output
66K
Released
2024-01-01
Knowledge cutoff
—
Retires
—
Open weights
yes
Input
text
Output
text

Price

US dollars per million tokens · most-agreed

Input
$0.28
Output
$0.42
Cache read
$0.14

Available from 2 hosts

HostIn $/MOut $/MCache rdCache wrContextOutputRetires
nano-gpt · deepseek-ai$0.28$0.42$0.14—164K66K—
qiniu-ai$0.2899$0.4348——128K32K—

More from DeepSeek

most-hosted first