#111 MoonshotAI: Kimi K2.6
none- Cost
- $0.020
- Time
- 127.4s
- Tokens
- 4,429 tok
Moonshot AI
Release: 2026-04-20
Tested on: 2026-05-22 00:18
moonshotai/kimi-k2.6::none
1T total (32B active)MoEWeights available
(medium)
(none)
Summary
Kimi K2.6 scores 5.6 on AI BENCHY and ranks #111. It has 8.3 reliability, a 38.3% pass rate, $0.088 total cost, and 13.86s average response time.
What makes Kimi K2.6 unique: It stands out most in Coding, where it ranks #4, while Anti-AI Tricks is its weakest area at #17.
Model facts
Researched on 2026-08-12
- Parameters
- 1T total (32B active)
- Architecture
- MoE
- Availability
- Weights available
- License
- Modified MIT
5.6
Consistency
9.2
8.3
$0.088
Total Output Tokens
16,563
Total Input Tokens
0
Input Price
$0.730 / 1M
Output Price
$3.490 / 1M
Flaky tests
2
Flaky tests had mixed outcomes across runs (at least one pass and one fail).
Hamster playing table tennis
Prompt: Create a detailed SVG illustration of a hamster playing table tennis.
Run history
| Tested on | Score | Reliability | Tests Correct | Total Cost | Compare |
|---|---|---|---|---|---|
| 2026-08-14 01:58 Re-test | 5.7 | 10.0 | $0.233 ↑ | Compare | |
| 2026-07-16 22:20 New test added | 5.8 | 10.0 | $0.233 ↑ | Compare | |
| 2026-06-04 13:36 New test added | 5.5 | 10.0 | $0.079 ↓ | Compare | |
| 2026-05-22 00:18 Suite changed | 5.6 | 8.3 | $0.088 | Current run | |
| 2026-04-23 10:54 First recorded run | 5.8 | N/A | $0.038 | Compare |
This run used a different benchmark suite. Keep suite changes in mind when reading historical movement.
Run comparison
| Run | Benchmark coverage | Score | Consistency | Reliability | Tests Correct | Flaky tests | Total Output Tokens | Total Input Tokens | Total Cost | Response Time (avg) |
|---|---|---|---|---|---|---|---|---|---|---|
| 2026-05-22 00:18 · Suite changed | 60/60 attempts | 5.6 | 9.2 | 8.3 | 7/20 | 2 | 16,563 | 0 | $0.088 | 13.86s |
| 2026-08-14 01:58 · Re-test | 66/66 attempts | 5.7 | 9.3 | 10.0 | 6/22 | 2 | 30,249 | 116,983 | $0.233 | 19.57s |
| Difference | — | -0.1 | -0.1 | -1.7 | +1 | 0 | -13686 | -116983 | -$0.145 | -5713ms |
Benchmark coverage differs: 60/60 attempts (Target: 3 repeats per test) versus 66/66 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.
Charts
Choose the first model, then click a second model to open a side-by-side page.
Top Models by Score
Score vs Total Cost
Response Time (avg)
Score vs Response Time (avg)
Total Output Tokens
Score vs Total Output Tokens
Quick Compare
Kimi K2.6nonevsGLM 5.1noneKimi K2.6nonevsQwen3.5-122B-A10BnoneKimi K2.6nonevsLing-3.0-flashnoneKimi K2.6nonevsKAT-Coder-Air V2.5highKimi K2.6nonevsInkling SmalllowKimi K2.6nonevsGemma 4 26B A4BnoneFree AvailableKimi K2.6nonevsGLM 5noneKimi K2.6nonevsKAT-Coder-Air V2.5mediumKimi K2.6nonevsNemotron 3 SupermediumFree AvailableKimi K2.6nonevsNemotron 3.5 LightninghighFree Available
Category Breakdown
| Category | Score | Consistency | Tests Correct |
|---|---|---|---|
| Anti-AI Tricks | 4.6 | 10.0 | |
| Coding | 6.8 | 9.8 | |
| Combined | 3.0 | 10.0 | |
| Data parsing and extraction | 10.0 | 10.0 | |
| Domain specific | 5.3 | 7.2 | |
| General Intelligence | 5.4 | 3.5 | |
| Instructions following | 6.5 | 10.0 | |
| Puzzle Solving | 3.2 | 9.8 | |
| Tool Calling | 10.0 | 10.0 | |
| Trivia | 3.0 | 10.0 |