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
VulcanBench Coding Intelligence Index v1. Every model the catalog carries a published VulcanBench CII v1 value for, ranked by that value.
A post-cutoff software-engineering benchmark with hidden functional tests and regression guards, reported for vendor coding-agent harnesses. 3 models in the ModelScale catalog have a published value.
Each point is a model with a published VulcanBench CII v1 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 VulcanBench CII v1 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 | Pass@1 with vendor coding-agent harnesses |
|---|---|---|---|
| 1 | Anthropic | 96.4 | |
| 2 | Anthropic | 89.2 | |
| 3 | OpenAI | 86.5 |
What the benchmark measures and how its value is published, from the source’s own record.
VulcanBench CII v1 (VulcanBench Coding Intelligence Index v1) is a coding benchmark published by CII v1 frontier results. A post-cutoff software-engineering benchmark with hidden functional tests and regression guards, reported for vendor coding-agent harnesses. Its evaluation set is described by the source as 38 validated post-cutoff repository tasks. The source rates its difficulty as mid-band frontier software engineering.
Each model's result is published by CII v1 frontier results as pass@1 with vendor coding-agent harnesses. ModelScale 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 CII v1 frontier results 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.