Navigate
AI BENCHY
Advertise here

Grok 4.20 (medium) vs Grok 4.3 (medium)

The average score is effectively tied at 7.1 vs 7.1. Grok 4.20 (medium) has the lower benchmark cost at $0.777 vs $0.779. Grok 4.20 (medium) is faster at 29.47s vs 47.45s, with pass rates of 63.6% vs 68.2%.

Last updated at: 2026-08-11

Rank
#101
Total Output Tokens
259,340
Response Time (avg)
29.47s
Total Cost
$0.777
Rank
#98
Total Output Tokens
241,421
Response Time (avg)
47.45s
Total Cost
$0.779
Recommended model Grok 4.20 (medium)

It has the best score here (7.1), while responding about 1.6x faster than Grok 4.3 (medium).

Detailed comparison

Metric Grok 4.20 Grok 4.20 medium Release: 2026-03-31 Grok 4.3 Grok 4.3 medium Release: 2026-05-01
Score 7.1 7.1
Rank #101 #98
Reliability 10.0 10.0
Consistency 8.5 8.6
Benchmark coverage 22/22 tests · 66/66 attempts 22/22 tests · 66/66 attempts
Tests Correct
Attempt pass rate 63.6% 68.2%
Flaky tests 4 4
Total Runs 66 66
Cost per result 9.709 5.990
Total Cost $0.777 $0.779
Input Price $1.250 / 1M $1.250 / 1M
Output Price $2.500 / 1M $2.500 / 1M
Total Input Tokens 102,791 140,031
Output Tokens 5,363 13,739
Reasoning Tokens 253,977 227,682
Response Time (avg) 29.47s 47.45s
Response Time (max) 199.66s 216.69s
Response Time (total) 648.35s 1043.83s

Model generation showcase

Hamster playing table tennis

Prompt: Create a detailed SVG illustration of a hamster playing table tennis.

#101 xAI: Grok 4.20

medium
Cost
$0.041
Time
110.3s
Tokens
16,336 tok

#98 xAI: Grok 4.3

medium
Cost
$0.009
Time
19.0s
Tokens
3,661 tok

Top Models by Score

Score vs Total Cost

Response Time (avg)

Score vs Response Time (avg)

Total Output Tokens

Score vs Total Output Tokens

Category Breakdown

Coding Score Consistency Attempt pass rate Flaky tests Tests Correct Response Time (avg) Input Tokens Output Tokens Reasoning Tokens
Grok 4.20 6.3 6.6 55.6% 1 109.93s 8,307 268 103,150
Grok 4.3 5.9 7.7 44.4% 1 41.23s 8,340 1,028 31,226

Quick Compare

Switch Comparison Pair

Muse Glimmer 30BlowvsGrok 4.3mediumMuse Glimmer 30BlowvsGrok 4.20mediumQwen3.7 PlusnonevsGrok 4.3mediumQwen3.7 PlusnonevsGrok 4.20mediumSolar Pro 4highvsGrok 4.3mediumSolar Pro 4highvsGrok 4.20mediumKAT-Coder-Pro V2.5highvsGrok 4.3mediumKAT-Coder-Pro V2.5highvsGrok 4.20mediumStep 3.7 FlashlowvsGrok 4.3mediumClaude Opus 4.8nonevsGrok 4.3mediumGemini 3.5 FlashnonevsGrok 4.20mediumStep 3.7 FlashlowvsGrok 4.20medium