readable AI benchmarks (simplified)

updated on: Nov 3, 2026, 7:17 AM UTC · 772 models

readable AI benchmarks

DeepSeek

DeepSeek V4.1 Flash 1017 (High)

reasoning · open weights · release date not listed

DeepSeek V4.1 Flash 1017 (High) has a Quality Score of 39.7, ranking 91st among 230 scored models. Per 1M tokens, pricing is $0.300 input, $1.20 output, and $0.0060 cached input. It has a 1M token context window. Its Reliability Score is 50.8, ranking 97th among 239 scored models, above average (total average: 46.3). Its Value Score is 33.7, high compared with the total Value average of 28.3.

Quality Score39.7coverage 100.0%: AA Intelligence and AA-Omniscience outcomes available
Value Score33.7
Factual reliability50.8
Cache Discount98%

Model specification

Reasoning
reasoning
Input modalities
none
Output modalities
none
Context window
1M tokens
Weights
open weights

Token prices USD per 1M tokens

Input
$0.300
Output
$1.20
Cached Input
$0.0060
Cache write
not available
Cost per task
not available

Capability

Intelligence
28.5
Omniscience (native net)
1.5
Correct
38.9%
Blended price
$0.184

Answer outcomes

Fully graded outcomes (Correct + Incorrect): 76.3%

Correct 38.9%Incorrect 37.4%Partial / not attempted 23.7%

Artificial Analysis benchmarks aa-v5

Current Intelligence Index components

AA-Briefcase v1.1
1415.7 Elo
AutomationBench-AA
68.5%
private 657-task split from dataset v1.0.6
GDP.pdf
14.2%
FrontierCode v2.0
16.8%
AA Stirrup · 132 tasks, 3 attempts per task
Terminal-Bench v4.0
28.8%
AA mini-swe-agent standardized runs · 66 tasks, pass@1 averaged over 3 repeats per task
AA-Omniscience v1.1
1.5 native net
HLE-Diamond
20.8%
CritPt v2 Hard
26.5%
AA Stirrup · hard subset, 40 of 70 problems
Terminal-Bench-Science v0.1.0
10.5%

Similar models

GraphOpen on Artificial Analysis

Support me! Patreon