Navigate
Advertise here

Kimi K3 (max) vs GPT-6 Luna (medium)

The average score is effectively tied at 7.9 vs 7.9. GPT-6 Luna (medium) has the lower benchmark cost at $0.031 vs $3.467. GPT-6 Luna (medium) is faster at 15.87s vs 142.84s, with pass rates of 77.3% vs 72.7%.

Last updated at: 2026-09-23

Compared models

Rank
#90
Total Output Tokens
224,128
Response Time (avg)
142.84s
Total Cost
$3.467
Rank
#94
Total Output Tokens
45,154
Response Time (avg)
15.87s
Total Cost
$0.031
Recommended model GPT-6 Luna (medium)

It has the best score here (7.9), while costing about 113.8x less than Kimi K3 (max).

Detailed comparison

Metric Kimi K3 Kimi K3 max Release: 2026-07-16 GPT-6 Luna GPT-6 Luna medium Release: 2026-09-23
Score 7.9 7.9
Rank #90 #94
Reliability 9.0 10.0
Consistency 9.2 8.9
Attempts 66/66 66/66
Tests Correct
Attempt pass rate 77.3% 72.7%
Flaky tests 2 3
Total Runs 66 66
Cost per result 21.668 0.218
Total Cost $3.467 $0.031
Input Price $3.000 / 1M $0.100 / 1M
Output Price $15.000 / 1M $0.500 / 1M
Total Input Tokens 34,960 78,946
Output Tokens 2,887 4,635
Reasoning Tokens 221,241 40,519
Response Time (avg) 142.84s 15.87s
Response Time (max) 766.58s 35.14s
Response Time (total) 2714.02s 349.09s
Parameters 2.8T total (104B active) ~400B total (~17B active)
Availability Weights available Closed

Model generation showcase

Hamster playing table tennis

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

#90 MoonshotAI: Kimi K3

max
Cost
$0.281
Time
506.9s
Tokens
18,863 tok

#94 GPT-6 Luna

medium
Cost
$0.002
Time
23.6s
Tokens
2,966 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
Kimi K3 10.0 10.0 100.0% 0 325.79s 8,061 460 84,012
GPT-6 Luna 6.6 4.6 77.8% 2 23.63s 7,302 438 10,222

Quick Compare

Switch Comparison Pair