readable AI benchmarks

DeepSeek

DeepSeek V4 Flash (Non-reasoning)

non-reasoning · open weights · Apr 24, 2026

Quality Score25.5
Value Score31.6
Reliability28.0
Cache Discount25%

Model specification

Reasoning
non-reasoning
Input modalities
text
Output modalities
text
Context window
1M tokens
Weights
open weights

Token prices USD per 1M tokens

Input
$0.094
Output
$0.195
Cached Input
$0.070
Cached Output
not available
Cost per task
not available

Capability

Intelligence
29.3
Coding
not available
Agentic
not available
Omniscience
-44.0
Correct
26.0%
Blended price
$0.087

Answer outcomes

Correct 26.0%Incorrect 70.0%Abstained 4.0%

Artificial Analysis benchmarks

SciCode
37%
Humanity’s Last Exam
8%
GPQA Diamond
72%
CritPt
0%
AA-Omniscience
28%
AA-LCR
37%

Similar models