Loading page…
Preferred languages
More languages supported by Google Translate
Open the selector to load the current language catalog.
Translations are provided by Google Translate.
Loading page…
Preferred languages
More languages supported by Google Translate
Open the selector to load the current language catalog.
Translations are provided by Google Translate.
Coding benchmark
Every model the catalog carries a published Kimi Code Bench v2 value for, ranked by that value.
A Moonshot AI internal coding-agent benchmark for realistic software-engineering tasks across mainstream programming languages and production technology stacks. 3 models in the ModelChoice catalog have a published value.
Each point is a model with a published Kimi Code Bench v2 value and complete token prices, on the 75% input / 25% output blend. Frontier models are spaced evenly along the cost axis.
3 models with a published Kimi Code Bench v2 value, ordered by that value, highest first. Models the source has not scored on this benchmark are not listed — they are unmeasured, not last.
| Rank | Model | Provider | Coding-agent pass rate |
|---|---|---|---|
| 1 | Moonshot AI | 77.1 | |
| 2 | Moonshot AI | 72.9 | |
| 3 | Moonshot AI | 62.0 |
What the benchmark measures and how its value is published, from the source’s own record.
Kimi Code Bench v2 is a coding benchmark published by Kimi K2.7 Code. A Moonshot AI internal coding-agent benchmark for realistic software-engineering tasks across mainstream programming languages and production technology stacks. Its evaluation set is described by the source as realistic coding-agent tasks. The source rates its difficulty as production software engineering.
Each model's result is published by Kimi K2.7 Code as coding-agent pass rate. ModelChoice prints that value to 1 decimal place, exactly as published, and never converts or rescales it. This page ranks the 3 models that carry a published value, highest value first; a model the source never scored is left out rather than ranked at zero. The source does not state whether a higher value is the better result, so neither does this page.
Evidence key: Observed
Rows are ordered by the value Kimi K2.7 Code published, highest first. The source does not state whether a higher value is the better result, so this page does not either: for a benchmark that measures a rate of failure, read the table from the bottom.