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
OTelBench (OpenTelemetry instrumentation tasks). Every model the catalog carries a published OTelBench value for, ranked by that value.
Quesma's OTelBench asks coding agents to add OpenTelemetry tracing, metrics and logging to small existing services in 11 programming languages. The score is the share of runs where the service then emits correct telemetry. 13 models in the ModelScale catalog have a published value.
Each point is a model with a published OTelBench value and complete token prices, on the 75% input / 25% output blend. Frontier models are spaced evenly along the cost axis.
13 models with a published OTelBench 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 | Agent edits a service in a Linux terminal, Harbor framework |
|---|---|---|---|
| 1 | Anthropic | 29.0 | |
| 2 | OpenAI | 26.1 | |
| 3 | Anthropic | 21.7 |
What the benchmark measures and how its value is published, from the source’s own record.
OTelBench (OTelBench (OpenTelemetry instrumentation tasks)) is an external benchmark published by Benchmarking OpenTelemetry: can AI trace your failed login? (Quesma). Quesma's OTelBench asks coding agents to add OpenTelemetry tracing, metrics and logging to small existing services in 11 programming languages. The score is the share of runs where the service then emits correct telemetry. Its evaluation set is described by the source as 23 OpenTelemetry instrumentation tasks in 11 languages, three attempts each (69 runs). The source rates its difficulty as distributed tracing and context propagation across services.
Each model's result is published by Benchmarking OpenTelemetry: can AI trace your failed login? (Quesma) as agent edits a service in a Linux terminal, Harbor framework. ModelScale prints that value to 1 decimal place, exactly as published, and never converts or rescales it. This page ranks the 13 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.
| 4 | 18.8 |
| 5 | 15.9 |
| 5 | OpenAI | 15.9 |
| 7 | OpenAI | 14.5 |
| 8 | Z.AI | 13.0 |
| 9 | DeepSeek | 11.6 |
| 9 | OpenAI | 11.6 |
| 11 | Anthropic | 5.8 |
| 12 | xAI | 4.3 |
| 13 | xAI | 2.9 |
Evidence key: Observed
Rows are ordered by the value Benchmarking OpenTelemetry: can AI trace your failed login? (Quesma) 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.