Browse › Natural Language Processing › Language Modelling › enwik8

Language Modelling archive 2025-07-28

enwik8 Benchmark (Language Modelling)

42 rows 40 with code listed 2 metrics

Syntology Since the archive (Syntology): none placed; 1 newer paper judged to report on this board so far.

A language model is a model of natural language. Language models are useful for a variety of tasks, including speech recognition, machine translation, natural language generation (generating more human-like text), optical character recognition, route optimization, handwriting recognition, grammar induction, and information retrieval.

Large language models (LLMs), currently their most advanced form, are predominantly based on transformers trained on larger datasets (frequently using words scraped from the public internet). They have superseded recurrent neural network-based models, which had previously superseded the purely statistical models, such as word n-gram language model.

Source: Wikipedia

The archive carries no text for this table; the description above is the archive's text for the task Language Modelling. archive 2025-07-28

Over time archive 2025-07-28

Direction inferred from the metric name, not from the archive: Number of params (lower is better). Not inferred (points only, no best-so-far line): Bit per Character (BPC). Points are placed at the row's paper date, not an evaluation date; 42 of 42 rows carry one.

The chart needs JavaScript; the table below carries every value.

Results archive 2025-07-28

Archive rows end at the archive snapshot, 2025-07-28: no result published after that date is in this table. Rank is the archive's row order at that snapshot; not re-ranked here. Metric values are the archive's strings. Column headers sort the table in your browser; each row keeps its archive rank.

Paper Code Ran Syntology Report
1 GPT-2 (48 layers, h=1600) 0.931542M ✓ Paper Code 2019 linked, not harvested report
2 Transformer-XL (24 layers, RMS dynamic eval, decay) 0.940277M ✓ Paper Code 2019 linked, not harvested report
3 Focus 0.94022M – Paper – 2023 no code linked report
4 Expire-Span (24 layers) 0.95208M – Paper Code 2021 linked, not harvested report
5 SRU++ Large 0.95195M – Paper Code 2021 linked, not harvested report
6 Feedback Transformer 0.9677M – Paper Code 2020 official: no sample here; runs from other or unrecorded repositories · 3 ran (of which 0 constructed an object rather than computing a result; 2 with no instrument failure: 0 honoured, 2 violated, 0 with no contract checked; 1 where Syntology's instrument failed) · 0 unverified report
7 Sandwich Transformer (adaptive span) 0.968209M – Paper Code 2019 linked, not harvested report
8 Compressive Transformer (24 layers) 0.97277M – Paper Code 2019 6 ran (of which 0 constructed an object rather than computing a result; 5 with no instrument failure: 0 honoured, 2 violated, 3 with no contract checked; 1 where Syntology's instrument failed) · 5 unverified report
9 Transformer-LS (large) 0.97110M – Paper Code 2021 official (archive's flag): 1 ran · 4 ran (of which 0 constructed an object rather than computing a result; 3 with no instrument failure: 0 honoured, 2 violated, 1 with no contract checked; 1 where Syntology's instrument failed) · 0 unverified report
10 SRU++ Base 0.97108M – Paper Code 2021 linked, not harvested report
11 Transformer (24 layers, 8k adaptive span) 0.98209M – Paper Code 2019 linked, not harvested report
12 Transformer-XL (24 layers) 0.99277M – Paper Code 2019 official (archive's flag): 8 ran · 67 ran (of which 37 constructed an object rather than computing a result; 49 with no instrument failure: 4 honoured, 1 violated, 44 with no contract checked; 18 where Syntology's instrument failed) · 76 unverified report
13 Longformer (30 layers, h=512) 0.99102M – Paper Code 2020 official (archive's flag): 9 ran · 22 ran (of which 4 constructed an object rather than computing a result; 14 with no instrument failure: 1 honoured, 1 violated, 12 with no contract checked; 8 where Syntology's instrument failed) · 13 unverified report
14 Sparse Transformer (30 layers, fixed attn) 0.9995M – Paper Code 2019 community repositories only · 5 ran (of which 4 constructed an object rather than computing a result; 4 with no instrument failure: 0 honoured, 0 violated, 4 with no contract checked; 1 where Syntology's instrument failed) · 1 unverified report
15 Routing Transformer (12 layers) 0.99 – Paper Code 2020 3 ran (of which 0 constructed an object rather than computing a result; 3 with no instrument failure: 1 honoured, 2 violated, 0 with no contract checked; 0 where Syntology's instrument failed) · 0 unverified report
16 Transformer-LS (small) 0.99 – Paper Code 2021 official (archive's flag): 1 ran · 4 ran (of which 0 constructed an object rather than computing a result; 3 with no instrument failure: 0 honoured, 2 violated, 1 with no contract checked; 1 where Syntology's instrument failed) · 0 unverified report
17 Hourglass 0.997 – Paper Code 2021 official (archive's flag): 1 ran · 3 ran (of which 0 constructed an object rather than computing a result; 3 with no instrument failure: 0 honoured, 2 violated, 1 with no contract checked; 0 where Syntology's instrument failed) · 1 unverified report
18 Longformer (12 layers, h=512) 1.0041M – Paper Code 2020 official (archive's flag): 9 ran · 22 ran (of which 4 constructed an object rather than computing a result; 14 with no instrument failure: 1 honoured, 1 violated, 12 with no contract checked; 8 where Syntology's instrument failed) · 13 unverified report
19 All-attention network (18 layers) 1.0139M – Paper Code 2019 5 ran (of which 0 constructed an object rather than computing a result; 5 with no instrument failure: 2 honoured, 3 violated, 0 with no contract checked; 0 where Syntology's instrument failed) · 0 unverified report
20 Transformer (12 layers, 8k adaptive span) 1.0239M – Paper Code 2019 linked, not harvested report
21 BP-Transformer (12 layers) 1.0238M – Paper Code 2019 official (archive's flag): 2 ran · 2 ran (of which 0 constructed an object rather than computing a result; 2 with no instrument failure: 0 honoured, 0 violated, 2 with no contract checked; 0 where Syntology's instrument failed) · 5 unverified report
22 Transformer+SSA 1.024 – Paper Code 2023 linked, not harvested report
23 Transformer-XL (18 layers) 1.0388M – Paper Code 2019 official (archive's flag): 8 ran · 67 ran (of which 37 constructed an object rather than computing a result; 49 with no instrument failure: 4 honoured, 1 violated, 44 with no contract checked; 18 where Syntology's instrument failed) · 76 unverified report
24 Skip Cross-Head Transformer-XL 1.03341M – Paper Code 2023 linked, not harvested report
25 Transformer (64 layers) 1.06235M – Paper Code 2018 linked, not harvested report
26 Transformer-XL (12 layers) 1.0641M – Paper Code 2019 official (archive's flag): 8 ran · 67 ran (of which 37 constructed an object rather than computing a result; 49 with no instrument failure: 4 honoured, 1 violated, 44 with no contract checked; 18 where Syntology's instrument failed) · 76 unverified report
27 SHA-RNN (4 layers, h=1024, attention head per layer) 1.06854M – Paper Code 2019 linked, not harvested report
28 SHA-RNN (4 layers, h=1024, single attention head) 1.07652M – Paper Code 2019 linked, not harvested report
29 64-layer Character Transformer Model 1.1144M – Paper Code 2018 linked, not harvested report
30 Mogrifier LSTM 1.14648M – Paper Code 2019 linked, not harvested report
31 LSTM 1.19548M – Paper Code 2019 linked, not harvested report
32 Cluster-Former (#C=512) 1.22 – Paper – 2020 no code linked report
33 AWD-LSTM (3 layers) 1.23247M – Paper Code 2018 community repositories only · 14 ran (of which 0 constructed an object rather than computing a result; 13 with no instrument failure: 2 honoured, 0 violated, 11 with no contract checked; 1 where Syntology's instrument failed) · 4 unverified report
34 Large mLSTM 1.2446M – Paper Code 2016 linked, not harvested report
35 Large FS-LSTM-4 1.2547M – Paper Code 2017 linked, not harvested report
36 Recurrent Highway Networks 1.2746M – Paper Code 2016 official: harvested, nothing ran · 1 ran (of which 0 constructed an object rather than computing a result; 0 with no instrument failure: 0 honoured, 0 violated, 0 with no contract checked; 1 where Syntology's instrument failed) · 4 unverified report
37 ByteNet 1.31 – Paper Code 2016 linked, not harvested report
38 LN HM-LSTM 1.3235M – Paper Code 2016 linked, not harvested report
39 SHA-LSTM (4 layers, h=1024, no attention head) 1.3351M – Paper Code 2019 linked, not harvested report
40 Hypernetworks 1.3427M – Paper Code 2016 3 ran (of which 0 constructed an object rather than computing a result; 1 with no instrument failure: 1 honoured, 0 violated, 0 with no contract checked; 2 where Syntology's instrument failed) · 1 unverified report
41 LSTM (7 layers) 1.67 – Paper Code 2013 31 ran (of which 0 constructed an object rather than computing a result; 28 with no instrument failure: 4 honoured, 0 violated, 24 with no contract checked; 3 where Syntology's instrument failed) · 6 unverified report
42 All-attention network (36 layers) 114M – Paper Code 2019 5 ran (of which 0 constructed an object rather than computing a result; 5 with no instrument failure: 2 honoured, 3 violated, 0 with no contract checked; 0 where Syntology's instrument failed) · 0 unverified report

All 42 rows shown. 42 link to a paper page on this site; 2 are marked as using additional training data in the archive. No GitHub stars are tracked; "Code" is the first repository the archive lists for the row. The archive carries no row tags, review links or community-submitted rows for this table; none are shown. archive 2025-07-28

Syntology Ran reads “N ran (of which C constructed an object rather than computing a result; K with no instrument failure: H honoured, V violated, P with no contract checked; I where Syntology's instrument failed) · U unverified”: of the code samples Syntology harvested from repositories linked to that row's paper (joined by arXiv id), N executed on a synthesized input and U are unverified (harvested, no recorded run). Of the N, C constructed an object rather than computing a result (a part of N, never taken away from it), K ran with no instrument failure (honoured or violated the contract Syntology drafted, or ran with no contract checked) and I are failures of Syntology's instrument, not of the code (a wrong contract draft, or a contract check that could not run). It counts code from repositories linked to that row's paper, not this result: the row's number was not reproduced and nothing here is a correctness claim. The other cell texts mean no graph line for the row: "linked, not harvested" (the archive links code, Syntology has not harvested it), "no code linked" (no code link in the archive), "not matched" (the row's paper URL matched no paper on this site). 19 rows have a graph line, from 14 distinct papers. Rows whose paper has a sample that ran: 19 (18 with a run with no instrument failure, 1 where every run was a failure of Syntology's instrument). Papers: 14 with a sample that ran: 13 with a run with no instrument failure, 1 where every run was a failure of Syntology's instrument. Whose code ran. When the archive marks a repository official for the row's paper, the cell starts with that repository's state (“official” is the archive's flag, not a verdict on who wrote the code): “official (archive's flag): N ran” counts the samples from it that ran, instrument failures included; “official: harvested, nothing ran”; when it has no sample for this paper and other samples ran, “community repositories only” if every one came from a repository the archive lists for the paper but neither marks official nor records as mentioned in it, or one Syntology harvested outside the archive's links, and “official: no sample here; runs from other or unrecorded repositories” if any came from a repository named in the paper or found in its text by Syntology, or has no repository recorded; “official: harvested for another paper” (no sample from it is linked to this paper) or “official: not harvested”. No state is printed when the archive marks none official. The cell's tooltip names the repositories the samples that ran came from. Of the 19 rows with a graph line, 13 have an archive-official repository: on 9 a sample from it ran (9 with a run with no instrument failure), on 2 the samples that ran came from community repositories only, and on 1 from other or unrecorded repositories. The filters above hide rows and never renumber them; the first two count, by default, only a run with no instrument failure, and offer any run, instrument failures included. Counting each paper once, of 285 samples: 169 ran (of which 45 constructed an object rather than computing a result; 132 with no instrument failure: 15 honoured, 15 violated, 102 with no contract checked; 37 where Syntology's instrument failed) · 116 unverified. Separately, 71 of those 285 are pointer-only (licence): the site points at that code rather than redistributing it, a licence property recorded for ran and unverified samples alike; each cell's tooltip carries the row's own pointer-only count. Syntology's record for this page last changed ; when this build read Syntology's graph is in the build record. Per-sample status is on the paper page; each Ran cell links to that paper's sample list. For agents: get_harvested_code_for_paper(arxiv_id) lists each paper's samples; how to connect.

Since the archive Syntology

Of the 8,888 newer papers whose tables Syntology has read (8,886 checked in full and 2 with a leaderboard still awaiting a verdict), 2 had this board among the leaderboards the extractor considered for their tables (it keeps at most 10 boards per paper and 4 per dataset; the lists of 1,596 of the 8,888 were already full without this board, so if it was crowded out of them it is not counted), and 1 proposed a result, which gave no cell that passed the extractor's own checks; none await a verdict and none were placed. Syntology has checked 8,886 of the 18,637 papers on this site that are newer than the archive (for 3,717 of them no archive leaderboard matched the paper's tables, so there was nothing further to check); 733 were read and have nothing to place (arXiv has no HTML version of the paper, or that version has no tables), 41 could not be read (the extractor's reply could not be parsed), and results from the other 8,977 appear after they are checked.

Newer papers whose tables Claude Sonnet 4.5 judged to report on this benchmark (a model's judgement, not a result; newest first by arXiv id)

Paper arXiv date Ran Syntology Outcome
Model of Models: When Does Emitting a Specialist Beat Attending, Adapting, or Tuning? arXiv:2608.21386 20 Jul 2026 9 harvested: 6 ran (of which 0 constructed an object rather than computing a result; 3 with no instrument failure: 1 honoured, 0 violated, 2 with no contract checked; 3 where Syntology's instrument failed) · 3 unverified refused by a rule: the table's column header does not name the leaderboard's metric

Syntology 1 newer paper judged to report on this board: the papers counted above as having proposed a result (0 placed); every one is listed here and in the JSON twin. The order is the arXiv identifier's, newest first, not a rank. The judgement is the model's, read from the paper's own tables; the outcome is what Syntology's checks did with it, and no value from a paper that was not placed is printed. “refused by a rule” names the extractor's rule; “no cell passed the extractor's checks” means the extractor's record holds no cell for this board, passed or refused. The 1 counted above as giving no cell that passed the extractor's own checks is the 1 marked “refused by a rule”. Drawn from 8,886 of 18,637 newer papers checked so far (733 more had nothing to place; 41 could not be read).

Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections