#147 MoonshotAI: Kimi K2.5
none- Cost
- $0.015
- Time
- 89.1s
- Tokens
- 5,421 tok
Moonshot AI
Release: 2026-01-27
Tested on: 2026-07-16 22:13
moonshotai/kimi-k2.5::none
1T total (32B active)MoEWeights available
(medium)
(none)
Summary
Kimi K2.5 scores 5.5 on AI BENCHY and ranks #147. It has 10.0 reliability, a 34.9% pass rate, $0.127 total cost, and 19.15s average response time.
What makes Kimi K2.5 unique: It stands out most in General Intelligence, where it ranks #1, while Anti-AI Tricks is its weakest area at #15.
Model facts
Researched on 2026-08-12
- Parameters
- 1T total (32B active)
- Architecture
- MoE
- Availability
- Weights available
- License
- Modified MIT
5.5
Consistency
8.6
10.0
Total Output Tokens
26,638
Total Input Tokens
89,322
Input Price
$0.571 / 1M
Output Price
$2.850 / 1M
Flaky tests
4
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 01:55 Re-test | 5.4 | 10.0 | $0.127 ↑ | Compare | |
| 2026-07-16 22:13 New test added | 5.5 | 10.0 | $0.127 ↑ | Current run | |
| 2026-06-04 13:32 New test added | 5.2 | 10.0 | $0.028 ↑ | Compare | |
| 2026-05-22 00:03 Suite changed | 5.3 | 10.0 | $0.026 | Compare | |
| 2026-04-20 17:48 First recorded run | 5.5 | N/A | $0.017 | 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) |
|---|---|---|---|---|---|---|---|---|---|---|
| 2026-07-16 22:13 · New test added | 66/66 attempts | 5.5 | 8.6 | 10.0 | 6/22 | 4 | 26,638 | 89,322 | $0.127 | 19.15s |
| 2026-04-20 17:48 · First recorded run | 54/54 attempts | 5.5 | 8.7 | N/A | 6/18 | 3 | 2,659 | 0 | $0.017 | 13.37s |
| Difference | — | 0.0 | -0.1 | 0 | +1 | +23979 | +89322 | +$0.111 | +5786ms |
Benchmark coverage differs: 66/66 attempts (Target: 3 repeats per test) versus 54/54 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 |
|---|---|---|
| 2026-06-04 15:40 | $0.400 / 1M | $1.900 / 1M |
| 2026-06-17 10:07 | $0.375 / 1M | $2.025 / 1M |
| 2026-07-16 15:36 | $0.571 / 1M | $2.850 / 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 K2.5nonevsMiMo-V2.5-PrononeKimi K2.5nonevsDeepSeek V4 Flash 0731noneKimi K2.5nonevsGPT-5.6 LunanoneKimi K2.5nonevsLaguna S 2.1highFree AvailableKimi K2.5nonevsKAT-Coder-Air V2.5lowKimi K2.5nonevsInkling SmallnoneKimi K2.5nonevsLaguna S 2.1mediumFree AvailableKimi K2.5nonevsLaguna XS 2.1noneFree AvailableKimi K2.5nonevsDeepSeek V4 Flash 0423noneKimi K2.5nonevsSeed-2.0-Codenone
Category Breakdown
| Category | Score | Consistency | Tests Correct |
|---|---|---|---|
| Anti-AI Tricks | 3.6 | 8.4 | |
| Coding | 5.5 | 10.0 | |
| Combined | 2.8 | 1.8 | |
| Data parsing and extraction | 7.3 | 5.8 | |
| Domain specific | 5.3 | 10.0 | |
| General Intelligence | 10.0 | 10.0 | |
| Instructions following | 6.5 | 10.0 | |
| Puzzle Solving | 3.0 | 10.0 | |
| Tool Calling | 10.0 | 10.0 | |
| Trivia | 3.0 | 10.0 |