readable AI benchmarks (simplified)

readable AI benchmarks

Upstage

Solar Mini 4

reasoning · closed weights · Sep 22, 2026

Solar Mini 4 does not yet have enough benchmark data for a Quality Score. Per 1M tokens, pricing is $0.100 input, $0.400 output, and $0.010 cached input. It accepts text input, outputs text, and has a 1,048,576 token context window. Its Reliability Score is 44.2, ranking 182nd among 288 scored models, below average (total average: 48.2).

Quality Scorenot availableQuality unavailable: missing AA Intelligence
Value Scorenot available
Factual reliability44.2
Cache Discount90%

Model specification

Reasoning
reasoning
Input modalities
text
Output modalities
text
Context window
1,048,576 tokens
Weights
closed weights

Token prices USD per 1M tokens

Input
$0.100
Output
$0.400
Cached Input
$0.010
Cache write
not available
Cost per task
not available

Capability

Intelligence
not available
Omniscience (native net)
-11.6
Correct
18.2%
Blended price
$0.067

Answer outcomes

Fully graded outcomes (Correct + Incorrect): 48.1%

Correct 18.2%Incorrect 29.9%Partial / not attempted 51.9%

Artificial Analysis benchmarks aa-v5

Current Intelligence Index components

AA-Briefcase v1.1
867.0 Elo
AutomationBench-AA
22.3%
private 657-task split from dataset v1.0.6
GDP.pdf
5.0%
Terminal-Bench v4.0
1.0%
AA mini-swe-agent standardized runs · 66 tasks, pass@1 averaged over 3 repeats per task
AA-Omniscience v1.1
-11.6 native net

Standalone evaluations

GDPval-AA v2.1
1092.1 Elo

Historical evaluations

SciCode v1.0.1
47.6%
AA Long Context Reasoning v1.1
83.3%
Humanity's Last Exam
25.8%
CritPt
1.4%

Similar models

Support me! Patreon