Best Long-Context LLMs

The benchmarks behind the long context ranking

Three public benchmarks decide this ranking. The heavier a benchmark's weight, the more it moves a model's score.

• 40%
  MRCR v2 (8-needle, 128K)

  Finds one of eight look-alike replies in a long chat

  Best on this testGemini 3.7 Flash97.0

• 35%
  AA-LCR

  Reasons across sets of long documents

  Best on this testKimi K388.7

• 25%
  LongBench v2

  Questions about very long texts

  Best on this testQwen3.8 Max Preview66.3

How the long context score is calculated

1. 1

   Rank on each benchmark

   Every model gets a percentile on each benchmark it has been measured on.

2. 2

   Steady the thin fields

   Where few models have taken a benchmark, that percentile is pulled toward the middle of the field.

3. 3

   Weigh and average

   The percentiles are averaged with the weights above into one score out of 100.

4. 4

   Qualify

   A model enters once it is measured on at least half the basket by weight, including one anchor benchmark.

372 models from 51 vendors have a long context score, each measured on at least 2 of the 3 benchmarks. Scores come from AA-graded results, official model cards and third-party evaluations.

Questions about the long context ranking