readable AI benchmarks (simplified)

readable AI benchmarks

SpaceXAI

Grok 4.20 0309 v2 (Reasoning)

reasoning · closed weights · Apr 7, 2026

Grok 4.20 0309 v2 (Reasoning) has a Quality Score of 48.4, ranking 96th among 551 scored models. Per 1M tokens, pricing is $1.25 input, $2.50 output, and $0.200 cached input. It has a 2M token context window. Its Reliability Score is 57.4, ranking 66th among 553 scored models, above average (total average: 35.6). Its Value Score is 35.8, high compared with the total Value average of 30.7.

Quality Score48.4coverage 100.0%: AA Intelligence and AA-Omniscience outcomes available
Value Score35.8
Factual reliability57.4
Cache Discount84%

Model specification

Reasoning
reasoning
Input modalities
none
Output modalities
none
Context window
2M tokens
Weights
closed weights

Token prices USD per 1M tokens

Input
$1.25
Output
$2.50
Cached Input
$0.200
Cache write
not available
Cost per task
not available

Capability

Intelligence
25.7
Coding
not available
Agentic
not available
Omniscience
14.8
Correct
27.5%
Blended price
$0.640

Answer outcomes

Fully graded outcomes (Correct + Incorrect): 40.1%

Correct 27.5%Incorrect 12.6%Partial / not attempted 59.9%

Artificial Analysis benchmarks

Humanity’s Last Exam
35%
GPQA Diamond
91%
CritPt
7%
AA-Omniscience
57%
AA-LCR
69%

Similar models

Support me! Patreon