Who leads the DROP benchmark?

MiniMax-Text-01 currently leads DROP with a score of 87.80 on f1.

What is the state-of-the-art score on DROP?

The state-of-the-art result on DROP is 87.80 (f1), achieved by MiniMax-Text-01 as of 2026.

How many models are tracked on DROP?

Codesota tracks 5 models on DROP.

When was the DROP leaderboard last updated?

The DROP leaderboard on Codesota includes results through 2026, with the earliest tracked result from 2021.

Codesota · Natural Language Processing · Question Answering · DROPTasks/Natural Language Processing/Question Answering

Question Answering · benchmark dataset · 2019 · EN

Discrete Reasoning Over Paragraphs.

Name: Discrete Reasoning Over Paragraphs Benchmark Results
Creator: Codesota
Published: 2021-01-01
License: https://creativecommons.org/licenses/by/4.0/

Reading-comprehension benchmark requiring arithmetic, counting, sorting, comparison, and other discrete reasoning over paragraphs.

Paper ↗Download dataset Submit a result ↵

§ 01 · Leaderboard

Best published scores.

5 results indexed across 1 metric. Shaded row marks current SOTA; ties broken by submission date.

Primary: f1 · higher is better

f1· primary

5 rows

#	Model	Org	Submitted	Paper / code	f1
01	MiniMax-Text-01	MiniMax	Jan 2025	MiniMax-01: Scaling Foundation Models with Lightning Att… · code	87.80
02	HRM-Text-1B	—	May 2026	pwc-dump · code	82.30
03	ByT5 XXL	—	May 2021	ByT5: Towards a token-free future with pre-trained byte-… · code	80
04	Apertus-70B-Instruct	—	Sep 2025	Apertus: Democratizing Open and Compliant LLMs for Globa… · code	50.80
05	OLMo-2-7B-1124 (olmOCR-peS2o)	—	Feb 2025	olmOCR: Unlocking Trillions of Tokens in PDFs with Visio… · code	43.70

Fig 2 · Rows sorted by score within each metric. Shaded row marks SOTA. Dates reflect model or paper release where available, otherwise the date Codesota accessed the source.

§ 03 · Progress

2 steps
of state of the art.

Each row below marks a model that broke the previous record on f1. Intermediate submissions are kept in the leaderboard above; only SOTA-setting entries are re-listed here.

Higher scores win. Each subsequent entry improved upon the previous best.

SOTA line · f1

May 28, 2021ByT5 XXL80
Jan 14, 2025MiniMax-Text-01MiniMax87.80

Fig 3 · SOTA-setting models only. 2 entries span May 2021 → Jan 2025.

§ 04 · Literature

4 papers
tied to this benchmark.

Every paper below corresponds to at least one row in the leaderboard above. Click through for the arXiv preprint and, when available, the reference implementation.

Apertus: Democratizing Open and Compliant LLMs for Global Language Environments
Sep 2025·Apertus-70B-Instruct
arXiv ↗Code
olmOCR: Unlocking Trillions of Tokens in PDFs with Vision Language Models
Feb 2025·OLMo-2-7B-1124 (olmOCR-peS2o)
arXiv ↗Code
MiniMax-01: Scaling Foundation Models with Lightning Attention
Jan 2025·MiniMax-Text-01
arXiv ↗Code
ByT5: Towards a token-free future with pre-trained byte-to-byte models
May 2021·ByT5 XXL
arXiv ↗Code

§ 06 · Contribute

Have a score that beats
this table?

Submit a checkpoint and a reproduction script. We will run it, publish the score, and — if it takes the top — annotate the step on the progress chart with your name.

Submit a result ↵Read submission guide

What a submission needs

01A public checkpoint or API endpoint
02A reproduction script with frozen commit + seed
03Declared evaluation environment (Python, deps)
04One row per metric declared by this dataset
05A contact so we can follow up on discrepancies

Discrete Reasoning Over Paragraphs.

Best published scores.

2 stepsof state of the art.

4 paperstied to this benchmark.

Neighbouring benchmarks.

Have a score that beatsthis table?

2 steps
of state of the art.

4 papers
tied to this benchmark.

Have a score that beats
this table?