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.
Reasoning benchmark
Every model the catalog carries a published AI-Needle value for, ranked by that value.
A long-context retrieval benchmark that measures whether a model can recover relevant information embedded deep inside very long contexts. 4 models in the ModelChoice catalog have a published value.
Each point is a model with a published AI-Needle value and complete token prices, on the 75% input / 25% output blend. Frontier models are spaced evenly along the cost axis.
4 models with a published AI-Needle 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 | Needle-in-a-haystack recall |
|---|---|---|---|
| 1 | Anthropic | 74 | |
| 2 | Alibaba | 68.7 | |
| 3 | Alibaba | 68.3 |
What the benchmark measures and how its value is published, from the source’s own record.
AI-Needle is a reasoning benchmark published by Qwen3.6 launch benchmarks. A long-context retrieval benchmark that measures whether a model can recover relevant information embedded deep inside very long contexts. Its evaluation set is described by the source as long-context retrieval. The source rates its difficulty as long-context memory.
Each model's result is published by Qwen3.6 launch benchmarks as needle-in-a-haystack recall. ModelChoice prints that value exactly as published, and never converts or rescales it. This page ranks the 4 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 | Z.AI | 63.3 |
Evidence key: Observed
Rows are ordered by the value Qwen3.6 launch benchmarks 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.