Benchmark Almanac

LongExtractionBench

Reducto Deep Extract 99.6% tops its leaderboard Claude Opus 4.8 70.7% leads the models we track Not in model cards

Extraction systems pull structured values from 225 long table-heavy documents averaging 358 pages and about 88,700 ground-truth fields each.

Compares four extraction platforms with GPT-5.5, Claude Opus 4.8 and Gemini 3.1 Pro called directly. Frontier models finished only about half the documents.

Made by micro1 Released Jun 2026 Documents, charts and text in images All tasks public Last model added May 2026 Scores refreshed 2 Nov 2026
Level
Expert Specialists with years of training, often over hours of work
Scored
Checked automatically A program checks the result against an answer key, unit tests or the expected final state.
Answers
Short answer A number, word or short string matched to a key.
micro1-research/longextract-bench 5 stars Paper Dataset Leaderboard
7
Entries on its own leaderboard, 3 models we track

Reducto Deep Extract tops the leaderboard with 99.6%

Best score per entry from the benchmark's own leaderboard, read on 2 Nov 2026. Where a model ran with several settings or harnesses, we keep its best. 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 Claude Opus 4.8, at 70.7%.

The newest model with a score here came out in May 2026, and nobody has tested a newer one. Among the 20 highest-ranked models on the capability index, these have no score yet:

  1. 99.6 Reducto Deep Extract
  2. 92.7 Extend MAX
  3. 77.5 LlamaExtract - Agentic
  4. 70.7 Claude Opus 4.8
  5. 52.7 GPT-5.5
  6. 48.6 Gemini 3.1 Pro
  7. 33.8 Datalab Extract - Balanced
  1. 1 Reducto Deep Extract 99.6%
  2. 2 Extend MAX 92.7%
  3. 3 LlamaExtract - Agentic 77.5%
  4. 4 Claude Opus 4.8 70.7%
  5. 5 GPT-5.5 52.7%
  6. 6 Gemini 3.1 Pro 48.6%
  7. 7 Datalab Extract - Balanced 33.8%
#EntryScoreSettingDateSource
1 Reducto Deep Extract 99.6% Official leaderboard
2 Extend MAX 92.7% Official leaderboard
3 LlamaExtract - Agentic 77.5% Official leaderboard
4 Claude Opus 4.8 70.7% May 2026 Official leaderboard
5 GPT-5.5 52.7% Apr 2026 Official leaderboard
6 Gemini 3.1 Pro 48.6% Feb 2026 Official leaderboard
7 Datalab Extract - Balanced 33.8% Official leaderboard

Number one over time

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

Anthropic
Own leaderboard Not tested since Jun 2026
Claude Opus 4.8 70.7%

Scores over time

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

On the benchmark’s own leaderboard Best so far
0% 20% 40% 60% 80% 100% 2026 Near the ceiling (85%) Claude Opus 4.8 GPT-5.5 Gemini 3.1 Pro 0% 20% 40% 60% 80% 100% 2026 Near the ceiling (85%) Claude Opus 4.8 GPT-5.5

No lab model card reports it yet

We read the headline results of 479 launch posts from 51 labs, and none of them list this benchmark.

More in Vision

All 234 in Vision
  • MMMU In use 11.5K college-level multimodal questions across 30 subjects. Multiple-choice and open accuracy. GPT-5.1 85.4%
  • MMMU-Pro In use Harder MMMU: college-level multimodal questions with more options and vision-only inputs to remove text shortcuts. Kimi K3.1 Pro 84.1%
  • CharXiv In use Chart understanding on real arXiv figures. Reasoning questions require synthesizing information from complex charts. Claude Fable 5 93.5%
  • MATH-Vision (MathVision) In use 3,040 competition math problems with visual context. Answer accuracy. Qwen 3.8 Max 97.7%
  • ChartQA In use Models answer human-written and generated questions about chart images that need visual and arithmetic reasoning. Scored by relaxed accuracy with a 5 percent numeric tolerance. Command A Vision 90.9%
  • DocVQA In use Models answer 50,000 questions about 12,000 scanned document images by reading their text and layout. Scoring is ANLS, a string similarity to reference answers. Seed1.5-VL 96.9%