Advertise here
#88

Kimi K3

Moonshot AI Release: 2026-07-16 Tested on: 2026-10-01 01:26 moonshotai/kimi-k3::max
2.8T total (104B active)MoEWeights available

Summary

Kimi K3 scores 8.3 on AI BENCHY and ranks #88. It has 9.3 reliability, a 78.3% pass rate, $3.409 total cost, and 139.41s average response time.

Model facts

Researched on 2026-08-12

Reported
Parameters
2.8T total (104B active)
Architecture
MoE
Availability
Weights available
License
Kimi K3
Parameter source 1 Parameter source 2 Availability source

Score

8.3

Consistency

9.3

Reliability

9.3

Total Cost (Current Price)

$3.409 ↓ -15.0%

Tested at: $4.013

Total Output Tokens

233,327

Total Input Tokens

170,700

Input Price

$0.830 / 1M

Output Price

$14.000 / 1M

Cache Read Price

$0.270 / 1M

Cache Write Price

N/A

Cache prices apply to input tokens. Reads reuse cached prompts; writes store them and can cost extra. Output tokens use the output price.

Tests Correct

Wrong Tests: 6

Attempt pass rate: 78.3%

Flaky tests

2

Flaky tests had mixed outcomes across runs (at least one pass and one fail).

Response Time (avg)

139.41s

Response Time (max): 766.58s

Response Time (total): 2788.26s

API error: 2 No answer: 2 Extra formatting: 1 Timed out: 1

Hamster playing table tennis

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

#88 MoonshotAI: Kimi K3

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

Run history

Tested on Score Reliability Tests Correct Total Cost Compare
2026-10-01 01:26 Re-test 8.3 9.3 $3.409 ↓ Current run
2026-07-16 22:18 New test added 8.0 9.1 $3.112 Compare
2026-07-16 17:10 Initial run 8.5 9.4 $3.112 Compare

This run used a different benchmark suite. Keep suite changes in mind when reading historical movement.

Price History

Historical pricing data for this model from OpenRouter.

Date Input Price Output Price Cache Read Price Cache Write Price
2026-08-14 12:51 $3.000 / 1M $15.000 / 1M N/A N/A
2026-10-06 14:54 $0.830 / 1M $14.000 / 1M $0.270 / 1M N/A

Charts

Choose the first model, then click a second model to open a side-by-side page.

Total Output Tokens

Score vs Total Output Tokens

Quick Compare

Kimi K3maxvsGemini 3.5 FlashhighKimi K3maxvsGPT-6 LunamediumKimi K3maxvsClaude Haiku 5.5highKimi K3maxvsClaude Opus 4.8mediumKimi K3maxvsGPT-5.6 TerrahighKimi K3maxvsLing 3.1 FlashmediumKimi K3maxvsGLM 5.2highKimi K3maxvsQwen3.7 PlusmediumKimi K3maxvsDeepSeek V4 Flash 0731highKimi K3maxvsGemini 4 Argonlow

Category Breakdown

Category Score Consistency Tests Correct
Agentic 10.0 10.0
Anti-AI Tricks 10.0 10.0
Coding 10.0 10.0
Combined 6.5 10.0
Data parsing and extraction 7.3 5.8
Domain specific 5.3 7.2
General Intelligence 10.0 10.0
Instructions following 10.0 10.0
Puzzle Solving 10.0 10.0
Tool Calling 3.0 10.0
Trivia 3.0 10.0

Compared models