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.
external benchmark
The Last Ones Cyber Range Completion Rate. Every model the catalog carries a published The Last Ones completion value for, ranked by that value.
Share of runs that completed the 32-step cyber range within the 100-million-token limit. 3 models in the ModelScale catalog have a published value.
Each point is a model with a published The Last Ones completion 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 The Last Ones completion 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 | Completion rate |
|---|---|---|---|
| 1 | OpenAI | 70 | |
| 2 | Anthropic | 60 | |
| 3 | Moonshot AI | 10 |
What the benchmark measures and how its value is published, from the source’s own record.
The Last Ones completion (The Last Ones Cyber Range Completion Rate) is an external benchmark published by UK AISI and CAISI Preliminary Assessment of Kimi K3's Cyber Capabilities. Share of runs that completed the 32-step cyber range within the 100-million-token limit. Its evaluation set is described by the source as 10 long-horizon runs. The source rates its difficulty as long-horizon cyber operations.
Each model's result is published by UK AISI and CAISI Preliminary Assessment of Kimi K3's Cyber Capabilities as completion rate. ModelScale prints that value to 0 decimal places, 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 UK AISI and CAISI Preliminary Assessment of Kimi K3's Cyber Capabilities 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.