AI BENCHY Compare

MoonshotAI: Kimi K3 vs OpenAI: GPT-5.4

Summary

The average score is effectively tied at 8.5 vs 8.5. GPT-5.4 (medium) has the lower benchmark cost at $1.210 vs $3.112. GPT-5.4 (medium) is faster at 22.35s vs 122.48s, with pass rates of 82.5% vs 76.2%.

Recommended modelGPT-5.4 (medium)It has the best score here (8.5), while costing about 2.6x less than Kimi K3 (max).

Last updated at: 2026-07-16

Metric	Kimi K3 Kimi K3 max Release: 2026-07-16	GPT-5.4 GPT-5.4 medium Release: 2026-03-05

Metric	Kimi K3 Kimi K3 max Release: 2026-07-16	GPT-5.4 GPT-5.4 medium Release: 2026-03-05
Score	8.5	8.5
Rank	#21	#23
Reliability	9.4	10.0
Consistency	9.2	8.6
Tests Correct
Attempt pass rate	82.5%	76.2%
Flaky tests	2	4
Total Runs	63	63
Cost per result	19.446	8.640
Total Cost	$3.112	$1.210
Input Price	$3.000 / 1M	$2.500 / 1M
Output Price	$15.000 / 1M	$15.000 / 1M
Total Input Tokens	34,916	34,108
Output Tokens	2,910	2,242
Reasoning Tokens	197,529	72,707
Response Time (avg)	122.48s	22.35s
Response Time (max)	766.58s	100.41s
Response Time (total)	2327.06s	469.29s

Generation showcase

Hamster playing table tennis

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

#21 MoonshotAI: Kimi K3

max

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

#23 GPT-5.4

medium

Cost: $0.214
Time: 199.6s
Tokens: 14,349 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

Category:

Anti-AI Tricks	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		10.24s	1,638	496	2,629
GPT-5.4	8.3	10.0	75.0%	0		4.11s	606	240	1,511

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-5.4	8.8	7.8	88.9%	1		44.36s	7,305	433	24,216

Combined	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		223.03s	12,792	399	20,460
GPT-5.4	10.0	10.0	100.0%	0		20.57s	11,019	301	3,543

Data parsing and extraction	Score	Consistency	Attempt pass rate	Flaky tests	Tests Correct	Response Time (avg)	Input Tokens	Output Tokens	Reasoning Tokens
Kimi K3	7.3	5.8	83.3%	1		16.72s	7,839	354	2,606
GPT-5.4	10.0	10.0	100.0%	0		5.32s	7,140	234	804

Domain specific	Score	Consistency	Attempt pass rate	Flaky tests	Tests Correct	Response Time (avg)	Input Tokens	Output Tokens	Reasoning Tokens
Kimi K3	5.9	7.2	55.6%	1		490.14s	794	80	83,180
GPT-5.4	5.3	7.2	44.4%	1		74.27s	619	61	34,748

General Intelligence	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		14.91s	732	191	871
GPT-5.4	4.7	3.1	33.3%	1		4.92s	477	145	321

Instructions following	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		7.66s	1,179	147	1,119
GPT-5.4	10.0	10.0	100.0%	0		3.11s	660	93	897

Puzzle Solving	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		7.36s	1,416	495	1,368
GPT-5.4	8.2	7.2	88.9%	1		9.14s	642	441	3,815

Tool Calling	Score	Consistency	Attempt pass rate	Flaky tests	Tests Correct	Response Time (avg)	Input Tokens	Output Tokens	Reasoning Tokens
Kimi K3	3.0	10.0	0.0%	0		0ms	0	0	0
GPT-5.4	10.0	10.0	100.0%	0		13.28s	5,445	264	1,031

Trivia	Score	Consistency	Attempt pass rate	Flaky tests	Tests Correct	Response Time (avg)	Input Tokens	Output Tokens	Reasoning Tokens
Kimi K3	3.0	10.0	0.0%	0		19.67s	465	288	1,284
GPT-5.4	3.0	10.0	0.0%	0		13.95s	195	30	1,821

Quick Compare

Switch Comparison Pair

Kimi K3maxvsGPT-5 Minimedium Kimi K3maxvsGLM 5medium Seed-2.0-LitemediumvsKimi K3max Kimi K3maxvsGPT-5.2 Chatnone Kimi K3maxvsStep 3.7 Flashmedium Kimi K3maxvsGLM 5.2medium Kimi K3maxvsGPT-5.2medium Claude Opus 4.7mediumvsKimi K3max Kimi K3maxvsGrok 4.5medium Muse Spark 1.1lowvsKimi K3max Muse Spark 1.1lowvsGPT-5.4medium DeepSeek V4 FlashhighvsGPT-5.4medium