readable AI benchmarks (simplified)

readable AI benchmarks

DeepSeek

DeepSeek V4 Pro

non-reasoning · open weights · Apr 24, 2026

DeepSeek V4 Pro has a Quality Score of 31.4, ranking 49th among 53 scored models. Per 1M tokens, pricing is $0.435 input, $0.870 output, and $0.0036 cached input. Its Reliability Score is 35.0, ranking 49th among 53 scored models, below average (total average: 50.1). Its Value Score is 32.8, low compared with the total Value average of 39.6.

Quality Score31.4coverage 78.8%: missing Epoch General ECI, Coding, CursorBench 3.2, Agentic and DeepSWE v1.1
Value Score32.8
Reliability35.0
Cache Discount99%

Model specification

Reasoning
non-reasoning
Input modalities
none
Output modalities
none
Context window
not available
Weights
open weights

Token prices USD per 1M tokens

Input
$0.435
Output
$0.870
Cached Input
$0.0036
Cache write
not available
Cost per task
not available

Capability

Intelligence
31.2
Coding
not available
Agentic
not available
Omniscience
-30.1
Correct
30.8%
Blended price
$0.177

Answer outcomes

Attempt rate: 91.6%

Correct 30.8%Incorrect 60.9%Abstained 8.4%

Artificial Analysis benchmarks

SciCode
42%
Humanity’s Last Exam
8%
GPQA Diamond
72%
CritPt
1%
AA-Omniscience
35%
AA-LCR
45%

Similar models