This is a large-scale collection of web-crawled documents in 198 world languages, produced by the HPLT project. The source of the data is Internet Archive and Common Crawl. For a detailed description of this and previous releases by HPLT, please refer to our website.

NB: the HPLT datasets are not hosted on HuggingFace! See download instructions below.

HPLT release v3.0

In July 2025, the European HPLT initiative has completed a new release of its monolingual datasets, offering better data quality, more annotations and metadata, and greatly increased volume. HPLT Monolingual Datasets 3.0 comprise some 50 terabytes of compressed data, covering 198 languages. More than half of the data represents the English language. Not counting the English majority portion, the dataset offers about 11.5 billion documents, 40 trillion Unicode characters, or 13.5 trillion tokens (using the Gemma 3 vocabulary). Overall, HPLT 3.0 is about three times larger than the previous release and likely constitutes the largest generally available multilingual dataset.

The dataset has been derived from some 7.2 petabytes of raw web crawls from the Internet Archive and the Common Crawl, spanning the period between 2012 and 2024. Text extraction from HTML documents was performed through the Trafilatura library, language identification with OpenLID 2.0, and deduplication, annotation, and filtering through the Monotextor pipeline.

Except quality and size, other distinguishing properties of the HPLT Monolingual Dataset is its sorting by a language-independent estimate of document quality and the rich annotations and metadata, including web register labels (for 104 of the languages in release 3.0), document- and segment-level language identification, annotation of personally identifiable information, and provenance information from the original crawl. Release 3.0 also fixes a deficiency in the Chinese data in the previous release, where double-width punctuation had been over-zealously normalized.

Except for Chinese, English, and Russian, each language-specific portion has been globally deduplicated.

Data processing was performed on dedicated storage and compute resources at the Czech and Norwegian national HPC infrastructures CESNET and Sigma2 NRIS, as well as on the EuroHPC LUMI system. The HPLT download site is hosted at the Sigma2 NIRD datalake.

New in this release compared to HPLT v2

Reflects substantially more raw web data, primarily from the Common Crawl
Additional metadata, including more information from the underlying crawl
Upgrade to Trafilatura 2.0 with empirical fine-tuning of extraction parameters
Plain-text and structured document representation, in simple, normalized XML
Better language identification; refined codes for Arabic and Chinese
Global deduplication for most languages; MinHash cluster size as metadata
Annotation with Turku web register labels for more than half the languages
Upgrade to newer, improved Web Docs Scorer (WDS) document quality estimates
Global sorting within each language by WDS and sharding into WDS bins (10–5)
Improved filtering for robots.txt opt-out, adult content, and credentials
Improved deduplication pipeline (global deduplication for most languages)

Downloading HPLT v3.0

For each language, the data is organized in smaller shards, sorted by document quality estimates (WDS). For Russian (in Cyrillic script), for example, the file rus_Cyrl/10_1.jsonl.zst is the first (and only) shard in the top WDS bin (scored as exactly 10), and rus_Cyrl/9_1.jsonl.zst…rus_Cyrl/9_103.jsonl.zstare the 103 shards in the bin for scores greater or equal to WDS9and less than10`.

The easiest way to download the data for a specific language is to use a command like wget -i with a language-specific mapping file containing full download addresses for all shards of this particular language, for example (for Crimean Tatar in Latin script):

wget -O - https://data.hplt-project.org/three/sorted/crh_Latn.map | wget -x -nH --cut-dirs=2 -i -

The above command retrieves the map for chr_Latn and feeds it as a list of download addresses into a second wget invocation, requesting the creation of local directories (-x), but cutting off the host and first two directory components (-nH --cut-dirs=2).

To download all available data, there is a larger mapping file for the full multilingual (excluding English) portion, amounting to a download of around 20 terabytes. The complete English data comprises some 30 terabytes and can be downloaded using its per-language mapping file. These can be retrieved using e.g. wget, and used as input directives for larger downloads, much like in the example above:

wget https://data.hplt-project.org/three/sorted/multilingual.map

wget https://data.hplt-project.org/three/sorted/eng_Latn.map

To speed up large downloads, it can be beneficial to use multiple parallel connections, for example using the --max-threads option in wget. We recommend to limit download parallelization to 16–32 threads, to avoid server-side rate limitations, which should allow download rates of around 250 gigabytes per hour.

Language-specific download links

Downloading with HuggingFace Datasets

To load a monolingual portion of the HPLT v3.0 dataset in the Huggingface Datasets format, you can run the following code to download the dataset files map and then load the .jsonl.zst files directly using load_datasets(). The Datasets package will then handle downloading of the files. If you would like to stream the files instead of downloading them all at once, set streaming=True within the load_dataset() function.

from datasets import load_dataset, Features, Value, Sequence, List
import requests

lang_code = "yor_Latn"  # Define your language-script code here, or "multilingual" for full multilingual portion minus English

r = requests.get(f"https://data.hplt-project.org/three/sorted/{lang_code}.map")

source_urls = r.text.strip().split("\n")

features = Features(
    {
        "f": Value("string"),
        "o": Value("int64"),
        "s": Value("int64"),
        "rs": Value("int64"),
        "u": Value("string"),
        "c": Value("string"),
        "ts": Value("timestamp[s]"),
        "de": Value("string"),
        "crawl_id": Value("string"),
        "lang": List(Value("string")),
        "prob": List(Value("float64")),
        "text": Value("string"),
        "xml": Value("string"),
        "html_lang": List(Value("string")),
        "cluster_size": Value("int64"),
        "seg_langs": List(Value("string")),
        "id": Value("string"),
        "filter": Value("string"),
        "pii": List(List(Value("int64"))),
        "doc_scores": List(Value("float64")),
        "web-register": {
            "MT": Value("float64"),
            "LY": Value("float64"),
            "SP": Value("float64"),
            "ID": Value("float64"),
            "NA": Value("float64"),
            "HI": Value("float64"),
            "IN": Value("float64"),
            "OP": Value("float64"),
            "IP": Value("float64"),
            "it": Value("float64"),
            "ne": Value("float64"),
            "sr": Value("float64"),
            "nb": Value("float64"),
            "re": Value("float64"),
            "en": Value("float64"),
            "ra": Value("float64"),
            "dtp": Value("float64"),
            "fi": Value("float64"),
            "lt": Value("float64"),
            "rv": Value("float64"),
            "ob": Value("float64"),
            "rs": Value("float64"),
            "av": Value("float64"),
            "ds": Value("float64"),
            "ed": Value("float64"),
            "cm": Value("float64"),
            "dp": Value("float64"),
            "dt": Value("float64"),
            "ib": Value("float64"),
            "oi": Value("float64"),
            "tr": Value("float64"),
            "ad": Value("float64"),
            "le": Value("float64"),
            "oo": Value("float64"),
            "ha": Value("float64"),
            "ma": Value("float64"),
            "on": Value("float64"),
            "pb": Value("float64"),
            "ss": Value("float64"),
            "tb": Value("float64"),
            "oe": Value("float64"),
            "pa": Value("float64"),
            "df": Value("float64"),
            "of": Value("float64"),
            "qa": Value("float64"),
            "rr": Value("float64"),
            "fh": Value("float64"),
            "ht": Value("float64"),
            "oh": Value("float64"),
            "ts": Value("float64"),
            "ol": Value("float64"),
            "po": Value("float64"),
            "pr": Value("float64"),
            "sl": Value("float64"),
            "fs": Value("float64"),
            "os": Value("float64"),
            "ta": Value("float64"),
            "tv": Value("float64")
        },
    }
)

ds = load_dataset("json", data_files=source_urls, features=features)

print(ds)

Statistics and validation

Summary statistics per language are available for download as a structured manifest.json, also including download links for the individual data files, per-language maps, and sample documents from various quality bins. Additionally, each language subdirectory provides compressed lists of unique domains, full URLs, and what are called normalized document signatures, together with their frequencies of occurence, for example nob_Latn/.domains.zst, nob_Latn/.urls.zst, and nob_Latn/.signatures.zst for Norwegian Bokmål.

Language	bytes	documents	segments	tokens	characters
ace_Arab	12749	7	72	6969	13861
ace_Latn	9.2329e+06	5225	149102	9.27194e+06	2.5397e+07
aeb_Arab	145577	177	2462	64086	177533
afr_Latn	3.87734e+09	2.13616e+06	5.66551e+07	2.68688e+09	8.80419e+09
als_Latn	1.2779e+10	1.11847e+07	1.62463e+08	1.00784e+10	2.77102e+10
amh_Ethi	1.24836e+09	571240	1.22545e+07	1.00555e+09	1.66535e+09
apc_Arab	229109	253	3494	88221	238181
arb_Arab	8.37614e+10	5.00711e+07	7.56572e+08	4.9566e+10	1.47182e+11
ars_Arab	2.06934e+06	1810	38430	968870	2.76278e+06
ary_Arab	1.96481e+07	17503	184651	1.07121e+07	3.27986e+07
arz_Arab	1.10897e+08	94125	1.12337e+06	6.23998e+07	1.76258e+08
asm_Beng	6.89823e+08	446306	6.54448e+06	4.79366e+08	1.14709e+09
ast_Latn	4.61011e+08	247531	5.0788e+06	3.08151e+08	1.0093e+09
awa_Deva	3.85411e+07	34188	354228	2.03499e+07	6.50232e+07
ayr_Latn	6.90081e+06	7449	120162	7.53512e+06	1.98012e+07
azb_Arab	1.83505e+08	94756	2.58497e+06	1.34707e+08	2.96101e+08
azj_Latn	1.51421e+10	1.10689e+07	2.44045e+08	1.59564e+10	4.12566e+10
bak_Cyrl	4.17521e+08	275718	3.96984e+06	3.93321e+08	8.0372e+08
bam_Latn	4.88638e+06	3638	64679	4.87638e+06	1.12839e+07
ban_Latn	3.91092e+07	16000	1.02105e+06	3.41714e+07	1.14841e+08
bel_Cyrl	5.67925e+09	2.99965e+06	5.5993e+07	4.0847e+09	1.01804e+10
bem_Latn	1.34493e+07	5344	142918	1.28912e+07	3.42116e+07
ben_Beng	3.71803e+10	2.55573e+07	3.59085e+08	1.63581e+10	6.25576e+10
bho_Deva	4.87399e+07	32789	473383	2.68768e+07	8.05295e+07
bjn_Arab	2.4971e+06	1306	30400	2.28464e+06	4.61119e+06
bjn_Latn	2.74946e+07	21227	364036	1.90762e+07	6.70367e+07
bod_Tibt	8.12033e+07	27863	481093	1.17868e+08	1.78277e+08
bos_Latn	4.82935e+10	3.70783e+07	6.41526e+08	3.20448e+10	9.92657e+10
bug_Latn	2.92831e+06	1173	32290	3.12132e+06	8.63034e+06
bul_Cyrl	7.7896e+10	4.29679e+07	9.78883e+08	4.89864e+10	1.45764e+11
cat_Latn	3.45719e+10	2.64118e+07	4.60848e+08	2.25402e+10	7.54288e+10
ceb_Latn	4.8916e+08	354235	6.77992e+06	3.84036e+08	1.25746e+09
ces_Latn	1.82916e+11	1.07802e+08	2.46849e+09	1.26248e+11	3.67843e+11
cjk_Latn	2.56179e+06	1081	29646	2.65488e+06	7.00067e+06
ckb_Arab	5.36151e+08	352126	4.98464e+06	4.72365e+08	9.56806e+08
cmn_Hans	5.40259e+12	2.21365e+09	6.02851e+10	2.97208e+12	4.13805e+12
cmn_Hant	2.45606e+11	1.13442e+08	2.37019e+09	1.47202e+11	1.95323e+11
crh_Latn	1.44888e+08	120315	1.52844e+06	1.28099e+08	3.15926e+08
cym_Latn	1.46134e+09	1.08198e+06	2.1101e+07	1.22767e+09	3.19176e+09
dan_Latn	8.73561e+10	5.24988e+07	1.33494e+09	6.27193e+10	2.08283e+11
deu_Latn	1.08506e+12	6.45363e+08	1.43806e+10	6.09313e+11	2.42507e+12
dik_Latn	2.65345e+06	1223	32639	3.32648e+06	6.66694e+06
dyu_Latn	3.00075e+06	1747	45172	3.49373e+06	7.36717e+06
dzo_Tibt	1.30172e+07	90	88550	2.05277e+07	1.98949e+07
ekk_Latn	2.63689e+10	1.37359e+07	4.25933e+08	2.06225e+10	6.06657e+10
ell_Grek	1.55946e+11	8.73908e+07	1.87499e+09	1.15569e+11	2.90056e+11
eng_Latn	3.25827e+13	1.80578e+10	4.35228e+11	1.62802e+13	7.2337e+13
epo_Latn	1.77994e+09	715290	2.3255e+07	1.24591e+09	3.73034e+09
eus_Latn	4.25142e+09	3.21948e+06	5.59265e+07	3.19136e+09	9.54646e+09
ewe_Latn	1.75704e+07	7137	218079	1.86855e+07	3.99497e+07
fao_Latn	3.21502e+08	323746	5.36184e+06	2.72706e+08	7.06501e+08
fij_Latn	2.07725e+07	12071	283590	2.10701e+07	5.93948e+07
fil_Latn	5.60269e+09	3.44124e+06	8.37043e+07	4.10511e+09	1.41188e+10
fin_Latn	9.38133e+10	4.95581e+07	1.36715e+09	7.39269e+10	2.19225e+11
fon_Latn	2.73227e+06	1469	24988	3.3823e+06	6.39682e+06
fra_Latn	1.0162e+12	6.03879e+08	1.56475e+10	5.84957e+11	2.27423e+12
fur_Latn	6.94786e+07	55016	1.11412e+06	7.08531e+07	2.14185e+08
fuv_Latn	1.44045e+07	9972	193399	1.49391e+07	3.49506e+07
gaz_Latn	1.01477e+08	63063	1.11134e+06	9.28262e+07	2.51115e+08
gla_Latn	2.71396e+08	204013	3.76981e+06	2.27705e+08	6.29982e+08
gle_Latn	1.25057e+09	786690	1.80665e+07	1.08783e+09	2.9638e+09
glg_Latn	5.31459e+09	4.03327e+06	6.65709e+07	3.11704e+09	1.1699e+10
gug_Latn	1.03362e+08	98974	1.69954e+06	7.43085e+07	2.41199e+08
guj_Gujr	5.16519e+09	3.46024e+06	4.67519e+07	3.33481e+09	8.38636e+09
hat_Latn	5.28153e+08	377114	7.86959e+06	4.04896e+08	1.18605e+09
hau_Latn	1.01531e+09	743843	1.5107e+07	7.97016e+08	2.35582e+09
heb_Hebr	4.79148e+10	2.60826e+07	6.47508e+08	3.70188e+10	7.91074e+10
hin_Deva	5.9972e+10	3.63252e+07	5.63696e+08	2.67672e+10	9.975e+10
hne_Deva	1.21755e+07	6322	95634	7.47182e+06	2.05146e+07
hrv_Latn	5.08172e+10	3.11567e+07	7.15451e+08	3.51469e+10	1.09108e+11
hun_Latn	1.35299e+11	7.51229e+07	1.77811e+09	1.02309e+11	2.95708e+11
hye_Armn	8.33672e+09	6.12071e+06	1.04487e+08	9.03616e+09	1.66412e+10
ibo_Latn	2.70718e+08	172837	4.01141e+06	2.5982e+08	6.03414e+08
ilo_Latn	5.53246e+07	43850	851053	4.44084e+07	1.3487e+08
ind_Latn	2.56153e+11	1.76108e+08	3.54427e+09	1.42115e+11	6.10519e+11
isl_Latn	6.84088e+09	4.29593e+06	9.34546e+07	6.14683e+09	1.56793e+10
ita_Latn	5.84823e+11	3.62986e+08	7.53909e+09	3.35456e+11	1.3016e+12
jav_Latn	3.71335e+08	239461	6.08345e+06	2.81119e+08	9.05496e+08
jpn_Jpan	1.60099e+12	6.67404e+08	3.57869e+10	8.76e+11	1.49653e+12
kab_Latn	2.43387e+07	15045	375223	2.14453e+07	4.90982e+07
kac_Latn	1.02018e+07	9032	149295	1.02602e+07	2.85214e+07
kam_Latn	1.49028e+06	1043	13065	1.76605e+06	3.63171e+06
kan_Knda	5.85547e+09	4.36419e+06	5.69032e+07	3.90595e+09	1.00064e+10
kas_Arab	2.09169e+06	1067	25454	1.75877e+06	3.55904e+06
kas_Deva	152450	66	571	147947	251926
kat_Geor	1.00526e+10	6.12929e+06	1.05893e+08	7.54849e+09	1.70096e+10
kaz_Cyrl	8.75866e+09	5.11957e+06	1.00639e+08	7.3411e+09	1.72129e+10
kbp_Latn	7.83331e+06	4774	68240	1.20192e+07	1.95624e+07
kea_Latn	3.50476e+06	3080	50813	2.43026e+06	7.27438e+06
khk_Cyrl	5.93504e+09	3.47859e+06	8.06166e+07	6.33191e+09	1.35331e+10
khm_Khmr	1.84434e+09	1.32366e+06	2.03944e+07	2.50309e+09	4.98338e+09
kik_Latn	7.93681e+06	8625	111888	7.77815e+06	1.79884e+07
kin_Latn	2.90616e+08	202519	3.73053e+06	2.54968e+08	6.93546e+08
kir_Cyrl	2.0203e+09	1.48906e+06	2.02656e+07	1.53549e+09	3.80122e+09
kmb_Latn	1.79382e+06	1178	20703	1.90339e+06	4.90187e+06
kmr_Latn	9.49185e+08	693889	1.20605e+07	7.91719e+08	1.96335e+09
knc_Arab	1.45563e+06	912	26736	1.51929e+06	2.33967e+06
knc_Latn	2.48346e+06	1387	30472	3.37973e+06	7.12004e+06
kor_Hang	1.50607e+11	7.47882e+07	2.33651e+09	9.75751e+10	1.64314e+11
ktu_Latn	7.76201e+06	4423	86549	8.17784e+06	2.24016e+07
lao_Laoo	1.56651e+08	87662	1.04696e+06	1.81446e+08	2.88255e+08
lij_Latn	1.64146e+07	6202	234666	1.53317e+07	3.0929e+07
lim_Latn	5.12171e+08	339706	6.56112e+06	3.71979e+08	1.1078e+09
lin_Latn	3.27883e+07	13561	441552	2.71829e+07	8.13272e+07
lit_Latn	3.59615e+10	2.04068e+07	5.11152e+08	2.87665e+10	8.07197e+10
lmo_Latn	1.38683e+08	116732	1.60841e+06	9.89455e+07	2.89874e+08
ltg_Latn	2.00932e+07	14140	218638	1.84417e+07	4.53443e+07
ltz_Latn	5.84169e+08	407481	7.96626e+06	4.33242e+08	1.33883e+09
lua_Latn	4.1414e+06	1634	50740	4.53105e+06	1.23946e+07
lug_Latn	5.28636e+07	49599	738304	4.69274e+07	1.23979e+08
luo_Latn	8.57884e+06	4611	103973	7.92437e+06	2.12114e+07
lus_Latn	4.14953e+08	294926	5.47319e+06	3.48683e+08	9.98553e+08
lvs_Latn	1.99862e+10	1.13234e+07	2.96733e+08	1.72447e+10	4.46281e+10
mag_Deva	2.43311e+06	513	75605	1.82292e+06	4.74567e+06
mai_Deva	7.05778e+07	28873	865082	4.43001e+07	1.16823e+08
mal_Mlym	1.07302e+10	8.15688e+06	9.01041e+07	6.64249e+09	1.90802e+10
mar_Deva	9.71831e+09	6.46084e+06	8.18362e+07	4.6767e+09	1.62002e+10
min_Latn	3.64913e+07	29395	596781	2.63115e+07	8.51086e+07
mkd_Cyrl	8.76829e+09	6.79145e+06	9.76109e+07	5.93043e+09	1.64243e+10
mlt_Latn	1.03778e+09	752735	1.72212e+07	9.81803e+08	2.46176e+09
mni_Beng	1.83794e+07	7573	189344	1.70622e+07	3.64384e+07
mos_Latn	4.5236e+06	1892	48001	5.81979e+06	1.16384e+07
mri_Latn	2.82025e+08	203007	4.12348e+06	2.3982e+08	6.85387e+08
mya_Mymr	4.00381e+09	1.97932e+06	3.68415e+07	4.24635e+09	7.21808e+09
nld_Latn	2.89043e+11	2.00689e+08	4.24748e+09	1.73413e+11	6.43031e+11
nno_Latn	2.2405e+09	1.50966e+06	3.19371e+07	1.58773e+09	4.92542e+09
nob_Latn	7.50453e+10	3.64871e+07	8.88906e+08	5.11551e+10	1.72132e+11
npi_Deva	8.86405e+09	6.21144e+06	7.62501e+07	4.87524e+09	1.50813e+10
nso_Latn	1.51224e+07	8183	234070	1.57664e+07	4.22513e+07
nus_Latn	528527	139	3278	766270	1.42048e+06
nya_Latn	2.68762e+08	177891	4.29311e+06	2.31979e+08	6.61407e+08
oci_Latn	1.66308e+08	106458	2.07166e+06	1.15326e+08	3.56397e+08
ory_Orya	1.40817e+09	1.29798e+06	9.43708e+06	1.53735e+09	2.21141e+09
pag_Latn	1.31976e+07	4496	171555	1.46059e+07	4.20388e+07
pan_Guru	2.51522e+09	1.51845e+06	2.21702e+07	2.31636e+09	4.27875e+09
pap_Latn	1.88525e+08	181779	2.41473e+06	1.36816e+08	4.64892e+08
pbt_Arab	1.40464e+09	918708	1.57483e+07	1.00765e+09	2.44663e+09
pes_Arab	2.7621e+11	1.32957e+08	4.06236e+09	1.81331e+11	5.49136e+11
plt_Latn	5.12107e+08	365680	6.75153e+06	4.3398e+08	1.23968e+09
pol_Latn	4.06616e+11	2.55893e+08	5.64384e+09	2.70101e+11	8.83717e+11
por_Latn	5.50084e+11	3.42528e+08	8.08785e+09	3.18853e+11	1.24275e+12
prs_Arab	3.55287e+09	2.45758e+06	4.75769e+07	2.00256e+09	6.39159e+09
quy_Latn	3.94001e+07	20199	565160	4.2765e+07	1.13999e+08
ron_Latn	1.52995e+11	9.59119e+07	2.16604e+09	1.02526e+11	3.39291e+11
run_Latn	2.11579e+08	235311	2.86853e+06	1.78848e+08	4.94911e+08
rus_Cyrl	8.59307e+12	3.30496e+09	1.00231e+11	4.40259e+12	1.595e+13
sag_Latn	4.62958e+06	2638	55772	5.1752e+06	1.40799e+07
san_Deva	2.78872e+08	59818	3.90726e+06	1.85157e+08	4.29315e+08
sat_Olck	7.28382e+06	4719	75497	1.1107e+07	1.16144e+07
scn_Latn	1.69874e+08	91611	2.039e+06	1.19946e+08	3.69055e+08
shn_Mymr	1.93813e+07	12287	157398	2.90415e+07	3.84962e+07
sin_Sinh	3.55884e+09	1.80283e+06	3.90258e+07	2.92338e+09	5.97592e+09
slk_Latn	5.69363e+10	3.63708e+07	7.68391e+08	4.02051e+10	1.16251e+11
slv_Latn	2.84362e+10	1.68121e+07	4.02193e+08	2.09247e+10	6.24579e+10
smo_Latn	2.32654e+08	161099	3.29799e+06	2.20451e+08	5.83679e+08
sna_Latn	2.62134e+08	183006	3.77958e+06	2.17012e+08	6.28522e+08
snd_Arab	5.97307e+08	363829	6.27533e+06	4.96112e+08	1.09322e+09
som_Latn	1.39533e+09	1.42382e+06	1.8755e+07	1.10301e+09	3.15936e+09
sot_Latn	2.39577e+08	152062	3.61913e+06	2.13919e+08	6.04692e+08
spa_Latn	1.20403e+12	7.2558e+08	1.63306e+10	6.58968e+11	2.753e+12
srd_Latn	8.12553e+07	66660	792085	5.79609e+07	1.75234e+08
srp_Cyrl	1.49864e+10	7.08171e+06	1.69868e+08	1.03789e+10	2.76843e+10
ssw_Latn	5.20474e+06	2789	94985	5.63088e+06	1.50385e+07
sun_Latn	2.78833e+08	185376	4.11052e+06	1.96535e+08	6.38091e+08
swe_Latn	1.68795e+11	9.77174e+07	2.47817e+09	1.11783e+11	3.75115e+11
swh_Latn	2.57702e+09	1.93591e+06	4.31838e+07	2.0635e+09	6.20961e+09
szl_Latn	4.18164e+07	30839	485518	3.16829e+07	8.40092e+07
tam_Taml	1.79397e+10	1.12662e+07	2.05225e+08	9.0269e+09	3.27291e+10
taq_Latn	2.7567e+06	827	48115	4.54676e+06	8.97492e+06
taq_Tfng	17837	5	171	27628	26799
tat_Cyrl	1.72649e+09	1.25906e+06	2.29245e+07	1.58331e+09	3.69134e+09
tel_Telu	8.89145e+09	6.23574e+06	8.19632e+07	5.55253e+09	1.50663e+10
tgk_Cyrl	3.95213e+09	2.57451e+06	4.18478e+07	3.33805e+09	7.9482e+09
tha_Thai	8.82601e+10	4.00081e+07	6.45136e+08	5.57641e+10	1.54505e+11
tir_Ethi	1.41116e+08	67624	1.24717e+06	1.38313e+08	1.91866e+08
tpi_Latn	1.94493e+07	12425	266506	1.93728e+07	5.83926e+07
tsn_Latn	1.8996e+07	9335	224171	1.92971e+07	5.35538e+07
tso_Latn	2.14762e+07	11680	292353	2.24503e+07	5.80403e+07
tuk_Latn	3.82449e+08	378448	4.86372e+06	3.70579e+08	9.02751e+08
tum_Latn	1.21203e+07	5654	157259	1.26477e+07	3.30254e+07
tur_Latn	2.28895e+11	1.59467e+08	3.11443e+09	1.49977e+11	5.12611e+11
twi_Latn	1.55538e+07	7896	213567	1.59479e+07	3.56205e+07
uig_Arab	1.04557e+09	645397	1.04776e+07	1.19735e+09	2.15813e+09
ukr_Cyrl	1.37162e+11	8.00269e+07	1.60566e+09	8.12156e+10	2.44658e+11
umb_Latn	4.38548e+06	2124	43464	4.46039e+06	1.20452e+07
urd_Arab	1.00517e+10	7.20656e+06	9.61188e+07	6.13055e+09	1.92455e+10
uzn_Latn	2.74834e+09	1.88281e+06	3.29448e+07	2.33642e+09	6.51135e+09
vec_Latn	1.32636e+08	102317	1.31702e+06	8.61756e+07	2.7615e+08
vie_Latn	2.34491e+11	1.45403e+08	3.5928e+09	1.42361e+11	4.75626e+11
war_Latn	1.03277e+07	9350	119247	8.00989e+06	2.54154e+07
wol_Latn	8.89604e+06	5056	161407	8.37268e+06	2.05158e+07
xho_Latn	3.59235e+08	253806	6.63664e+06	3.27797e+08	8.6304e+08
ydd_Hebr	3.52683e+08	162585	4.30668e+06	3.60245e+08	7.15206e+08
yor_Latn	2.54982e+08	171248	3.62308e+06	2.30281e+08	5.59874e+08
yue_Hant	4.72524e+08	217261	4.61984e+06	2.13832e+08	2.77005e+08
zgh_Tfng	4.09375e+06	3488	34992	6.60944e+06	6.55445e+06
zsm_Latn	3.05097e+10	1.73653e+07	5.03869e+08	1.82977e+10	7.12286e+10
zul_Latn	4.50878e+08	336440	8.01597e+06	4.10539e+08	1.12129e+09

The counts of documents per language or total storage sizes in the above statistics could be used to approximately validate each language sub-directory, but for more thorough validation of individual data files or full downloads, MD5 checksum files are provided with naming conventions parallel to the data and per-language map files. For example: nob_Latn/.10_1.jsonl.md5 for the first data file in Norwegian Bokmål, and nob_Latn.md5 for its full set of data files.

License and takedown

License

These data are released under this licensing scheme:

We do not own any of the text from which these text data has been extracted.*
We license the actual packaging of these text data under the Creative Commons CC0 license ("no rights reserved").

Notice and take down policy

Notice: Should you consider that our data contains material that is owned by you and should therefore not be reproduced here, please:

Clearly identify yourself, with detailed contact data such as an address, telephone number or email address at which you can be contacted.
Clearly identify the copyrighted work claimed to be infringed.
Clearly identify the material that is claimed to be infringing and information reasonably sufficient to allow us to locate the material.
You can reach us at hplt-datasets@ufal.mff.cuni.cz

Take down: We will comply to legitimate requests by removing the affected sources from the next release of the corpora.

It is your responsibility that any use of the data complies with any applicable legal framework, such as, among others, the EU Copyright Directive 2019/790 and the General Data Protection Regulation 2018, as amended.

Cite us

@misc{oepen2025hplt30largescalemultilingual,
      title={HPLT 3.0: Very Large-Scale Multilingual Resources for LLM and MT. Mono- and Bi-lingual Data, Multilingual Evaluation, and Pre-Trained Models}, 
      author={Stephan Oepen and Nikolay Arefev and Mikko Aulamo and Marta Bañón and Maja Buljan and Laurie Burchell and Lucas Charpentier and Pinzhen Chen and Mariya Fedorova and Ona de Gibert and Barry Haddow and Jan Hajič and Jindřich Helcl and Andrey Kutuzov and Veronika Laippala and Zihao Li and Risto Luukkonen and Bhavitvya Malik and Vladislav Mikhailov and Amanda Myntti and Dayyán O'Brien and Lucie Poláková and Sampo Pyysalo and Gema Ramírez Sánchez and Janine Siewert and Pavel Stepachev and Jörg Tiedemann and Teemu Vahtola and Dušan Variš and Fedor Vitiugin and Tea Vojtěchová and Jaume Zaragoza},
      year={2025},
      eprint={2511.01066},
      archivePrefix={arXiv},
      primaryClass={cs.CL},
      url={https://arxiv.org/abs/2511.01066}, 
}

Funding

This project has received funding from the European Union’s Horizon Europe research and innovation programme under grant agreement No 101070350 and from UK Research and Innovation (UKRI) under the UK government’s Horizon Europe funding guarantee (grant number 10052546)

Downloads last month: 183

Models trained or fine-tuned on HPLT/HPLT3.0