#114 MoonshotAI: Kimi K2.7 Code
medium- Cost
- $0.025
- Time
- 138.0s
- Tokens
- 6,093 tok
Moonshot AI
Release: 2026-06-12
Tested on: 2026-08-14 02:05
moonshotai/kimi-k2.7-code::medium
1T total (32B active)MoEWeights available Summary
Kimi K2.7 Code scores 7.4 on AI BENCHY and ranks #114. It has 10.0 reliability, a 63.6% pass rate, $0.709 total cost, and 80.09s average response time.
What makes Kimi K2.7 Code unique: It stands out most in Coding, where it ranks #3, while Puzzle Solving is its weakest area at #16.
Model facts
Researched on 2026-08-12
- Parameters
- 1T total (32B active)
- Architecture
- MoE
- Availability
- Weights available
- License
- Modified MIT
7.4
Consistency
8.0
10.0
Total Output Tokens
239,546
Total Input Tokens
72,085
Input Price
$0.710 / 1M
Output Price
$3.500 / 1M
Flaky tests
5
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 02:05 Re-test | 7.4 | 10.0 | $0.709 ↓ | Current run | |
| 2026-07-16 22:20 New test added | 7.5 | 10.0 | $0.690 ↓ | Compare | |
| 2026-06-12 18:35 Initial run | 7.0 | 9.2 | $0.669 | 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) |
|---|---|---|---|---|---|---|---|---|---|---|
| 2026-08-14 02:05 · Current run | 66/66 attempts | 7.4 | 8.0 | 10.0 | 11/22 | 5 | 239,546 | 72,085 | $0.709 | 80.09s |
| 2026-07-16 22:20 · New test added | 66/66 attempts | 7.5 | 8.3 | 10.0 | 12/22 | 4 | 262,507 | 72,073 | $0.690 | 84.25s |
| Difference | — | -0.1 | -0.3 | 0.0 | -1 | +1 | -22961 | +12 | +$0.020 | -4160ms |
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 |
|---|---|---|
| 2026-06-12 18:36 | $0.950 / 1M | $4.000 / 1M |
| 2026-06-17 10:07 | $0.740 / 1M | $3.500 / 1M |
| 2026-07-09 20:59 | $0.720 / 1M | $3.490 / 1M |
| 2026-07-14 10:48 | $0.719 / 1M | $3.490 / 1M |
| 2026-07-16 15:36 | $0.750 / 1M | $3.500 / 1M |
| 2026-07-18 15:17 | $1.000 / 1M | $4.400 / 1M |
| 2026-07-20 15:56 | $0.850 / 1M | $3.800 / 1M |
| 2026-07-21 16:10 | $0.820 / 1M | $3.750 / 1M |
| 2026-07-25 03:22 | $0.780 / 1M | $3.500 / 1M |
| 2026-07-28 08:40 | $0.730 / 1M | $3.500 / 1M |
| 2026-08-05 23:12 | $0.700 / 1M | $3.500 / 1M |
| 2026-08-12 20:39 | $0.670 / 1M | $3.400 / 1M |
| 2026-08-14 12:51 | $0.710 / 1M | $3.500 / 1M |
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 K2.7 CodemediumvsQwen3.7 MaxnoneKimi K2.7 CodemediumvsGPT-5.6 LunamediumKimi K2.7 CodemediumvsGemini 3.1 Flash Lite PreviewmediumKimi K2.7 CodemediumvsLongCat 2.0mediumKimi K2.7 CodemediumvsGemini 3.8 FlashlowKimi K2.7 CodemediumvsQwen3.8 2.4T A95BhighKimi K2.7 CodemediumvsGrok Build 0.1mediumKimi K2.7 CodemediumvsGrok 4.6lowKimi K2.7 CodemediumvsGPT-5.4 MinimediumKimi K2.7 CodemediumvsGemini 3 Flash Previewlow
Category Breakdown
| Category | Score | Consistency | Tests Correct |
|---|---|---|---|
| Anti-AI Tricks | 7.3 | 5.8 | |
| Coding | 7.8 | 9.3 | |
| Combined | 7.3 | 5.8 | |
| Data parsing and extraction | 10.0 | 10.0 | |
| Domain specific | 3.7 | 6.5 | |
| General Intelligence | 10.0 | 10.0 | |
| Instructions following | 9.9 | 10.0 | |
| Puzzle Solving | 5.9 | 7.7 | |
| Tool Calling | 3.0 | 10.0 | |
| Trivia | 3.0 | 10.0 |