MRCR v2

Reported in recent model launches

These model reports include this benchmark. Their harnesses, task subsets, scoring policies, and resource budgets can differ; follow each report’s evaluation notes before joining the numbers.

What it measures

Long-context tasks require tracking multiple references and retrieving the requested information from a large input.

How to read the result

Compare performance at the same context length and number of competing references. The hardest setting can expose failures hidden by an overall average.

Limits and comparison traps

Needle counts and token ranges identify distinct slices. A near-perfect result at one range does not establish performance at the full advertised context limit.

Benchmark reference notes

Multi-needle retrieval and coreference disambiguation among many similar interleaved conversations up to 1M tokens.

Where this fits in an evaluation suite

A large context window is a capacity specification. Successful retrieval and cross-document reasoning are observed capabilities. Test both at the actual lengths your workflow needs.

Start with a capability hypothesis, then choose a second test with a different failure mode. Keep the checkpoint, reasoning effort, agent tools, sampling policy, and scoring procedure fixed when measuring a training intervention. Record failures and missing outputs in the denominator.

A reproducible comparison

Pin the benchmark release, task split, model configuration, tools, sampling budget, and evaluation harness described by the primary source.

Publish the exact run configuration, item coverage, score units, and available uncertainty. If a source supplies only an aggregate or a single run, preserve that limit. A new model name or a rounded score alone does not establish a capability improvement.