Benchmark Almanac

CroSS-MR

GPT-4-Turbo (snapshot unspecified) 57.8% tops its leaderboard Qwen1.5-72B 54.6% leads the models we track
Not in model cards

Models read a scenario from a novel and select the reason for a character’s decision. The score is the percentage of correct choices.

The MemRole team ran all 445 questions. MemRole-32B scored 69.83% accuracy across the 126 novels.

Released
April 2024
Kind
Reading comprehension
Level
School
Scored
Checked automatically
Answers
Multiple choice
Tasks
All public
Measures
Capability
Language
English
In and out
Text → Text
Interaction
Single turn
Refresh
One fixed set
Run by
Anyone
Last model added
February 2024
Scores refreshed
November 21st, 2026
9
Entries in its paper, 4 models we track

GPT-4-Turbo (snapshot unspecified) tops the leaderboard with 57.8% Qwen1.5-72B leads with 54.6% among models released by February 2024

Best score per entry from the benchmark's paper, read on November 21st, 2026. A paper’s table doesn’t add newer models. Where a model ran with several settings or harnesses, we keep its best. Every run is in the next chart. Names in gray aren't in our model list: fine-tunes, agent systems, products and older models. They don't count toward leaders. The best model we track is Qwen1.5-72B, at 54.6%. This view shows only models we track, each at its best.

The newest model with a score here came out in February 2024, and nobody has tested a newer one. Among the 20 highest-ranked models on the capability index, these have no score yet:

  1. 1 GPT-4-Turbo (snapshot unspecified) Incremental Updating 57.8%
  2. 2 Qwen1.5-72B Hierarchical Merging 54.6%
  3. 3 Mixtral-8x7B-MoE Incremental Updating 52.1%
  4. 4 GPT-3.5-Turbo (snapshot unspecified) Hierarchical Merging 51.7%
  5. 5 Mistral-7B-Instruct-v0.2 Hierarchical Merging 50.6%
  6. 6 Qwen1.5-14B-Chat Incremental Updating 47.4%
  7. 7 vicuna-13b-v1.5-16k Incremental Updating 46.3%
  8. 8 vicuna-7b-v1.5-16k Hierarchical Merging 44.9%
  9. 9 Qwen1.5-7B-Chat Incremental Updating 44.5%

Source: The CroSS-MR paper

#EntryScoreSettingDateSource
1 GPT-4-Turbo (snapshot unspecified) 57.8% Incremental Updating Official leaderboard
2 Qwen1.5-72B 54.6% Hierarchical Merging Feb 2024 Official leaderboard
3 Mixtral-8x7B-MoE 52.1% Incremental Updating Official leaderboard
4 GPT-3.5-Turbo (snapshot unspecified) 51.7% Hierarchical Merging Official leaderboard
5 Mistral-7B-Instruct-v0.2 50.6% Hierarchical Merging Dec 2023 Official leaderboard
6 Qwen1.5-14B-Chat 47.4% Incremental Updating Feb 2024 Official leaderboard
7 vicuna-13b-v1.5-16k 46.3% Incremental Updating Official leaderboard
8 vicuna-7b-v1.5-16k 44.9% Hierarchical Merging Official leaderboard
9 Qwen1.5-7B-Chat 44.5% Incremental Updating Feb 2024 Official leaderboard
#ModelScoreSettingDateSource
1 Qwen1.5-72B 54.6% Hierarchical Merging Feb 2024 Official leaderboard
2 Mistral-7B-Instruct-v0.2 50.6% Hierarchical Merging Dec 2023 Official leaderboard
3 Qwen1.5-14B-Chat 47.4% Incremental Updating Feb 2024 Official leaderboard
4 Qwen1.5-7B-Chat 44.5% Incremental Updating Feb 2024 Official leaderboard

Scores at each effort level

Models the leaderboard ran at two or more effort levels. We don’t have enough cost data here to say what the extra effort costs.

Lowest effort to highest, light to dark
Qwen1.5-72B
Incremental… Hierarchica…
54.6%
Mistral-7B-Instruct-v0.2
Incremental… Hierarchica…
50.6%
Qwen1.5-14B-Chat
Incremental…
47.4%
Qwen1.5-7B-Chat
Hierarchica…
44.5%

Source: The CroSS-MR paper

ModelEffortHarnessScoreCost per taskOutput tokens
Qwen1.5-72B Hierarchical Merging 54.6%
Qwen1.5-72B Incremental Updating 52.4%
Mistral-7B-Instruct-v0.2 Hierarchical Merging 50.6%
Mistral-7B-Instruct-v0.2 Incremental Updating 48.3%
Qwen1.5-14B-Chat Incremental Updating 47.4%
Qwen1.5-14B-Chat Hierarchical Merging 47.4%
Qwen1.5-7B-Chat Incremental Updating 44.5%
Qwen1.5-7B-Chat Hierarchical Merging 44%

Scores at each effort level

Models the leaderboard ran at two or more effort levels. We don’t have enough cost data here to say what the extra effort costs.

Lowest effort to highest, light to dark
Qwen1.5-72B
Incremental… Hierarchica…
54.6%
Mistral-7B-Instruct-v0.2
Incremental… Hierarchica…
50.6%
Qwen1.5-14B-Chat
Incremental…
47.4%
Qwen1.5-7B-Chat
Hierarchica…
44.5%

Source: The CroSS-MR paper

ModelEffortHarnessScoreCost per taskOutput tokens
Qwen1.5-72B Hierarchical Merging 54.6%
Qwen1.5-72B Incremental Updating 52.4%
Mistral-7B-Instruct-v0.2 Hierarchical Merging 50.6%
Mistral-7B-Instruct-v0.2 Incremental Updating 48.3%
Qwen1.5-14B-Chat Incremental Updating 47.4%
Qwen1.5-14B-Chat Hierarchical Merging 47.4%
Qwen1.5-7B-Chat Incremental Updating 44.5%
Qwen1.5-7B-Chat Hierarchical Merging 44%

Number one over time

Lab claims count from the day the card came out, independent runs from the model’s release.

Qwen
Benchmark paper Not tested since Apr 2024
Qwen1.5-72B 54.6%

Source: The CroSS-MR paper

Scores over time

All the scores we have, one dot each, so a model tested at several settings appears more than once.

Plot with other benchmarks
In the benchmark’s paper Best so far
0% 20% 40% 60% 80% 100% 2024 Near the ceiling (85%) Qwen1.5-72B Mistral-7B-Instruct-v0.2 0% 20% 40% 60% 80% 100% 2024 Near the ceiling (85%) Qwen1.5-72B

Source: Official CroSS-MR leaderboard

No lab model card reports it yet

We read the headline results of 540 launch posts from 57 labs, and none of them list this benchmark.

More in Knowledge

All 332 in Knowledge
  • MMLU-Pro In use Harder 10-option version of MMLU with reasoning-focused questions across 14 subjects. Intern-S2-397B 89.8%
  • MMLU In use 57-subject multiple-choice exam questions from elementary to professional level. Scored by accuracy. Seed1.8 92.3%
  • SuperGPQA In use 26,529 graduate-level multiple-choice questions across 285 disciplines. Accuracy. Qwen 3.6 Max (Preview) 73.9%
  • WideSearch In use 200 bilingual broad info-seeking tasks: agent collects many atomic facts into a table. Scored by success rate and row/item F1. Tencent Hy4 preview 83.9%
  • BBEH micro average In use Models solve harder replacements for each of the 23 BIG-Bench Hard tasks, 4,520 examples in total. The score is the share of all examples answered correctly. Gemma 4 31B IT 74.4%
  • HLE-Diamond In use Models answer 1,000 cleaned expert questions from Humanity's Last Exam, 500 on reasoning and 500 on knowledge, closed book. Graded on exact answers. Claude Fable 5.2 77.9%

Data audit

9 of 10 checks pass, 1 warning. Look first at the Newest score check. The newest came 2 years ago, in February 2024, for Qwen1.5-14B-Chat. Every benchmark

Catalog record 2 pass
  • Pass Task access All tasks are public.
  • Pass Confidence Checked against a primary source.
Scores 1 warning 3 pass 1 note
  • Pass Independent runs Scores from the benchmark’s paper (9 entries).
  • Pass Models scored 4 models we track, plus 5 other entries on its leaderboard.
  • Warning Newest score The newest came 2 years ago, in February 2024, for Qwen1.5-14B-Chat.
  • Note Frontier models None of its models are on the capabilities index.
  • Pass Headroom The best is 54.6%, under 85%, so it still separates models.
Leaderboard import 3 pass
  • Pass Importer We read the board on aclanthology.org: 8 rows for models we track and 5 other entries at the last read.
  • Pass Last read November 21st, 2026.
  • Pass Read worked The last read worked.
Consistency 1 pass 1 note
  • Note Lab claims No model has both a lab number and an independent run.
  • Pass Known issues No published audit found problems.

How we check