Advertise here
#196

Kimi K2.6

Moonshot AI Release: 2026-04-20 Tested on: 2026-10-01 01:26 moonshotai/kimi-k2.6::medium
1T total (32B active)MoEWeights available
(medium) (none)

Summary

Kimi K2.6 scores 6.8 on AI BENCHY and ranks #196. It has 10.0 reliability, a 60.9% pass rate, $1.594 total cost, and 121.25s average response time.

What makes Kimi K2.6 unique: It stands out most in Instructions following, where it ranks #1, while Puzzle Solving is its weakest area at #17.

Model facts

Researched on 2026-08-12

Reported
Parameters
1T total (32B active)
Architecture
MoE
Availability
Weights available
License
Modified MIT
Parameter source Availability source

Score

6.8

Consistency

8.4

Reliability

10.0

Total Cost (Current Price)

$1.594 ↑ +8.1%

Tested at: $1.474

Total Output Tokens

344,502

Total Input Tokens

226,880

Input Price

$0.950 / 1M

Output Price

$4.000 / 1M

Cache Read Price

$0.160 / 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: 11

Attempt pass rate: 60.9%

Flaky tests

4

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

Response Time (avg)

121.25s

Response Time (max): 876.20s

Response Time (total): 2667.55s

Wrong answer: 4 Timed out: 3 Did not follow instructions: 2 Invalid tool call: 1 No answer: 1

Hamster playing table tennis

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

#196 MoonshotAI: Kimi K2.6

medium
Cost
$0.013
Time
103.4s
Tokens
3,620 tok

Run history

Tested on Score Reliability Tests Correct Total Cost Compare
2026-10-01 01:26 Re-test 6.8 10.0 $1.594 ↑ Current run
2026-07-16 22:59 New test added 7.2 9.4 $1.222 ↑ Compare
2026-06-04 13:59 New test added 7.2 7.8 $0.891 ↓ Compare
2026-05-22 00:14 Suite changed 7.4 8.3 $0.916 Compare
2026-04-23 10:54 First recorded run 7.7 N/A $0.722 Compare

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

Run comparison

RunBenchmark coverageScoreConsistencyReliabilityTests CorrectFlaky testsTotal Output TokensTotal Input TokensTotal CostResponse Time (avg)
2026-10-01 01:26 · Current run69/69 attempts6.88.410.012/234344,502226,880$1.594121.25s
2026-05-22 00:14 · Suite changed60/60 attempts7.48.38.312/204331,8770$0.91654.11s
Difference—-0.6+0.1+1.700+12625+226880+$0.678+67139ms

Benchmark coverage differs: 69/69 attempts (Target: 3 repeats per test) versus 60/60 attempts (Target: 3 repeats per test). Totals and repeat-sensitive metrics are not directly comparable.

These two runs used different benchmark suites, so the deltas reflect both model changes and suite changes.

Price History

Historical pricing data for this model from OpenRouter.

Date Input Price Output Price Cache Read Price Cache Write Price
2026-06-04 15:40 $0.684 / 1M $3.420 / 1M N/A N/A
2026-06-10 13:42 $0.680 / 1M $3.410 / 1M N/A N/A
2026-06-29 01:00 $0.660 / 1M $3.410 / 1M N/A N/A
2026-06-30 21:25 $0.550 / 1M $3.200 / 1M N/A N/A
2026-07-08 21:00 $0.650 / 1M $3.410 / 1M N/A N/A
2026-07-25 03:22 $0.646 / 1M $2.720 / 1M N/A N/A
2026-08-01 01:38 $0.600 / 1M $3.410 / 1M N/A N/A
2026-08-04 12:41 $0.589 / 1M $2.480 / 1M N/A N/A
2026-08-07 22:44 $0.580 / 1M $2.440 / 1M N/A N/A
2026-08-14 12:51 $0.950 / 1M $4.000 / 1M N/A N/A
2026-10-03 09:03 $0.950 / 1M $4.000 / 1M $0.160 / 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 K2.6mediumvsClaude Opus 4.8noneKimi K2.6mediumvsQwen3.5-FlashmediumKimi K2.6mediumvsClaude Sonnet 5.5lowKimi K2.6mediumvsInklinglowFree AvailableKimi K2.6mediumvsGPT-5.6 LunalowKimi K2.6mediumvsMercury 2.5 PreviewmediumKimi K2.6mediumvsGemma 4 31BmediumFree AvailableKimi K2.6mediumvsGPT-6 SolnoneKimi K2.6mediumvsEmber-1highKimi K2.6mediumvsSpace Bunny Alphamax

Category Breakdown

Category Score Consistency Tests Correct
Agentic 5.0 10.0
Anti-AI Tricks 7.0 8.0
Coding 5.7 8.6
Combined 6.9 5.9
Data parsing and extraction 10.0 10.0
Domain specific 5.3 7.2
General Intelligence 10.0 10.0
Instructions following 10.0 10.0
Puzzle Solving 6.0 7.4
Tool Calling 10.0 10.0
Trivia 3.0 10.0

Compared models