modelbenchmark.io

DeepSeek V4.1 Flash 1017 Abliterated

DeepSeek · deepseek-v4-1-flash-1017-abliterated

Compare

NEAR AI's refusal-removed DeepSeek V4.1 Flash 1017 for security research, with text-only reasoning and tool use inside a private TEE

Specification

most-agreed values

Context
1M
Max output
384K
Released
2026-10-29
Knowledge cutoff
2025-05
Retires
—
Open weights
yes
Input
text
Output
text

Price

US dollars per million tokens · most-agreed

Input
$0.30
Output
$1.20
Cache read
$0.006

Available from 2 hosts

HostIn $/MOut $/MCache rdCache wrContextOutputRetires
hopper$0.30$1.20$0.03—1M1M—
nearai · deepseek-ai$0.30$1.20$0.006—1M384K—

More from DeepSeek

most-hosted first