Benchmark Almanac

LMentry

davinci-002 66.1% leads since November 2022
Not in model cards

Models perform simple language tasks such as identifying words with a given letter or choosing the longer word. Task-specific rules check the answers.

Made by
Avia Efrat and colleagues
Released
November 2022
Kind
Language reasoning
Level
Everyday
Scored
Checked automatically
Answers
Short answer
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
March 2022
Scores refreshed
November 20th, 2026
16
Entries in its paper, 9 models we track

davinci-002 leads with 66.1% among models released by March 2022 davinci-002 leads with 66.1% among models released by March 2022

Best score per entry from the benchmark's paper, read on November 20th, 2026. A paper’s table doesn’t add newer models. 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. This view shows only models we track, each at its best.

The newest model with a score here came out in March 2022, and nobody has tested a newer one. Among the 20 highest-ranked models on the capability index, these have no score yet:

  1. 1 davinci-002 66.1%
  2. 2 InstructGPT 175B 48.4%
  3. 3 Tk-dpne 11B 41%
  4. 4 T0pp 11B 38.3%
  5. 5 Tk-dp 11B 37.2%
  6. 6 Tk-d 11B 35.8%
  7. 7 T0p 11B 35.7%
  8. 8 InstructGPT 6B 34.3%
  9. 9 T0 11B 31.6%
  10. 10 GPT-3 175B (davinci) 29.9%
  11. 11 InstructGPT 1.3B 24.8%
  12. 12 GPT-3 6.7B 21.8%
  13. 13 T5-LM 11B 20.6%
  14. 14 InstructGPT 350M 20.4%
  15. 15 GPT-3 XL (babbage) 19.7%
  16. 16 GPT-3 Medium 15.4%

Source: The LMentry paper

#EntryScoreSettingDateSource
1 davinci-002 66.1% Mar 2022 Official leaderboard
2 InstructGPT 175B 48.4% Jan 2022 Official leaderboard
3 Tk-dpne 11B 41% Official leaderboard
4 T0pp 11B 38.3% Official leaderboard
5 Tk-dp 11B 37.2% Official leaderboard
6 Tk-d 11B 35.8% Official leaderboard
7 T0p 11B 35.7% Official leaderboard
8 InstructGPT 6B 34.3% Jan 2022 Official leaderboard
9 T0 11B 31.6% Official leaderboard
10 GPT-3 175B (davinci) 29.9% May 2020 Official leaderboard
11 InstructGPT 1.3B 24.8% Jan 2022 Official leaderboard
12 GPT-3 6.7B 21.8% Jun 2020 Official leaderboard
13 T5-LM 11B 20.6% Official leaderboard
14 InstructGPT 350M 20.4% Jan 2022 Official leaderboard
15 GPT-3 XL (babbage) 19.7% Jun 2020 Official leaderboard
16 GPT-3 Medium 15.4% Jun 2020 Official leaderboard
#ModelScoreSettingDateSource
1 davinci-002 66.1% Mar 2022 Official leaderboard
2 InstructGPT 175B 48.4% Jan 2022 Official leaderboard
3 InstructGPT 6B 34.3% Jan 2022 Official leaderboard
4 GPT-3 175B (davinci) 29.9% May 2020 Official leaderboard
5 InstructGPT 1.3B 24.8% Jan 2022 Official leaderboard
6 GPT-3 6.7B 21.8% Jun 2020 Official leaderboard
7 InstructGPT 350M 20.4% Jan 2022 Official leaderboard
8 GPT-3 XL (babbage) 19.7% Jun 2020 Official leaderboard
9 GPT-3 Medium 15.4% Jun 2020 Official leaderboard

Rankings by part

We read 25 parts from the benchmark’s paper, each ranking models on one part of the test, such as sentence containing, sentence not containing and word containing. davinci-002 leads 22 of them, and InstructGPT 175B leads the other 3. Arrows show how far a model moved from its overall rank among the models we track.

sentence containing

9 models we track have a score here.

  1. 1 davinci-002 96.1%
  2. 2 InstructGPT 175B 95.5%
  3. 3 GPT-3 175B (davinci) ↑1 90.4%
  4. 4 InstructGPT 6B ↓1 87.5%
  5. 5 InstructGPT 350M ↑2 81.8%
  6. 6 InstructGPT 1.3B ↓1 79.9%
  7. 7 GPT-3 XL (babbage) ↑1 75.5%
  8. 8 GPT-3 6.7B ↓2 74.3%
  9. 9 GPT-3 Medium 72%

Number one over time

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

OpenAI
Benchmark paper Not tested since Nov 2022
davinci-002 66.1%

Source: The LMentry 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% 2020 2021 2022 Near the ceiling (85%) davinci-002 InstructGPT 175B GPT-3 175B (davinci) 0% 20% 40% 60% 80% 100% 2020 2021 2022 Near the ceiling (85%) davinci-002 InstructGPT 175B

Source: Official LMentry 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 4 years ago, in March 2022, for davinci-002. 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 (16 entries).
  • Pass Models scored 9 models we track, plus 7 other entries on its leaderboard.
  • Warning Newest score The newest came 4 years ago, in March 2022, for davinci-002.
  • Note Frontier models None of its models are on the capabilities index.
  • Pass Headroom The best is 66.1%, under 85%, so it still separates models.
Leaderboard import 3 pass
  • Pass Importer We read the board on aclanthology.org: 9 rows for models we track and 7 other entries at the last read.
  • Pass Last read November 20th, 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