Nllb 200 Distilled 1.3B Ct2 Int8

🚀 Fast-Inference with Ctranslate2

This project speeds up inference and reduces memory usage by 2x - 4x through int8 inference in C++ on CPU or GPU. It's a quantized version of facebook/nllb-200-distilled-1.3B.

✨ Features

• Language Support:
  • Supported languages include ace, acm, acq, aeb, af, ajp, ak, als, am, apc, ar, ars, ary, arz, as, ast, awa, ayr, azb, azj, ba, bm, ban, be, bem, bn, bho, bjn, bo, bs, bug, bg, ca, ceb, cs, cjk, ckb, crh, cy, da, de, dik, dyu, dz, el, en, eo, et, eu, ee, fo, fj, fi, fon, fr, fur, fuv, gaz, gd, ga, gl, gn, gu, ht, ha, he, hi, hne, hr, hu, hy, ig, ilo, id, is, it, jv, ja, kab, kac, kam, kn, ks, ka, kk, kbp, kea, khk, km, ki, rw, ky, kmb, kmr, knc, kg, ko, lo, lij, li, ln, lt, lmo, ltg, lb, lua, lg, luo, lus, lvs, mag, mai, ml, mar, min, mk, mt, mni, mos, mi, my, nl, nn, nb, npi, nso, nus, ny, oc, ory, pag, pa, pap, pbt, pes, plt, pl, pt, prs, quy, ro, rn, ru, sg, sa, sat, scn, shn, si, sk, sl, sm, sn, sd, so, st, es, sc, sr, ss, su, sv, swh, szl, ta, taq, tt, te, tg, tl, th, ti, tpi, tn, ts, tk, tum, tr, tw, tzm, ug, uk, umb, ur, uzn, vec, vi, war, wo, xh, ydd, yo, yue, zh, zsm, zu.
  • Language details: ace_Arab, ace_Latn, acm_Arab, acq_Arab, aeb_Arab, afr_Latn, ajp_Arab, aka_Latn, amh_Ethi, apc_Arab, arb_Arab, ars_Arab, ary_Arab, arz_Arab, asm_Beng, ast_Latn, awa_Deva, ayr_Latn, azb_Arab, azj_Latn, bak_Cyrl, bam_Latn, ban_Latn,bel_Cyrl, bem_Latn, ben_Beng, bho_Deva, bjn_Arab, bjn_Latn, bod_Tibt, bos_Latn, bug_Latn, bul_Cyrl, cat_Latn, ceb_Latn, ces_Latn, cjk_Latn, ckb_Arab, crh_Latn, cym_Latn, dan_Latn, deu_Latn, dik_Latn, dyu_Latn, dzo_Tibt, ell_Grek, eng_Latn, epo_Latn, est_Latn, eus_Latn, ewe_Latn, fao_Latn, pes_Arab, fij_Latn, fin_Latn, fon_Latn, fra_Latn, fur_Latn, fuv_Latn, gla_Latn, gle_Latn, glg_Latn, grn_Latn, guj_Gujr, hat_Latn, hau_Latn, heb_Hebr, hin_Deva, hne_Deva, hrv_Latn, hun_Latn, hye_Armn, ibo_Latn, ilo_Latn, ind_Latn, isl_Latn, ita_Latn, jav_Latn, jpn_Jpan, kab_Latn, kac_Latn, kam_Latn, kan_Knda, kas_Arab, kas_Deva, kat_Geor, knc_Arab, knc_Latn, kaz_Cyrl, kbp_Latn, kea_Latn, khm_Khmr, kik_Latn, kin_Latn, kir_Cyrl, kmb_Latn, kon_Latn, kor_Hang, kmr_Latn, lao_Laoo, lvs_Latn, lij_Latn, lim_Latn, lin_Latn, lit_Latn, lmo_Latn, ltg_Latn, ltz_Latn, lua_Latn, lug_Latn, luo_Latn, lus_Latn, mag_Deva, mai_Deva, mal_Mlym, mar_Deva, min_Latn, mkd_Cyrl, plt_Latn, mlt_Latn, mni_Beng, khk_Cyrl, mos_Latn, mri_Latn, zsm_Latn, mya_Mymr, nld_Latn, nno_Latn, nob_Latn, npi_Deva, nso_Latn, nus_Latn, nya_Latn, oci_Latn, gaz_Latn, ory_Orya, pag_Latn, pan_Guru, pap_Latn, pol_Latn, por_Latn, prs_Arab, pbt_Arab, quy_Latn, ron_Latn, run_Latn, rus_Cyrl, sag_Latn, san_Deva, sat_Beng, scn_Latn, shn_Mymr, sin_Sinh, slk_Latn, slv_Latn, smo_Latn, sna_Latn, snd_Arab, som_Latn, sot_Latn, spa_Latn, als_Latn, srd_Latn, srp_Cyrl, ssw_Latn, sun_Latn, swe_Latn, swh_Latn, szl_Latn, tam_Taml, tat_Cyrl, tel_Telu, tgk_Cyrl, tgl_Latn, tha_Thai, tir_Ethi, taq_Latn, taq_Tfng, tpi_Latn, tsn_Latn, tso_Latn, tuk_Latn, tum_Latn, tur_Latn, twi_Latn, tzm_Tfng, uig_Arab, ukr_Cyrl, umb_Latn, urd_Arab, uzn_Latn, vec_Latn, vie_Latn, war_Latn, wol_Latn, xho_Latn, ydd_Hebr, yor_Latn, yue_Hant, zho_Hans, zho_Hant, zul_Latn.
• Tags: ctranslate2, int8, float16, nllb, translation
• License: cc-by-nc-4.0
• Datasets: flores-200
• Metrics: bleu, spbleu, chrf++

📦 Installation

To install the necessary package, run the following command:

pip install ctranslate2

📚 Documentation

Compatibility

The checkpoint is compatible with ctranslate2>=3.22.0.

• Use compute_type=int8_float16 for device="cuda".
• Use compute_type=int8 for device="cpu".

Conversion Process

The model was converted on 2023 - 11 - 30 using CTranslate2==3.22.0 with the following code:

from ctranslate2.converters import TransformersConverter
TransformersConverter(
    "facebook/nllb-200-distilled-1.3B",
    activation_scales=None,
    copy_files=['tokenizer.json', 'generation_config.json', 'README.md', 'special_tokens_map.json', 'tokenizer_config.json', '.gitattributes'],
    load_as_float16=True,
    revision=None,
    low_cpu_mem_usage=True,
    trust_remote_code=True,
).convert(
    output_dir=str(tmp_dir),
    vmap = None, 
    quantization="int8",
    force = True,
)

📄 License

This is just a quantized version. License conditions are intended to be identical to the original Hugging Face repo.

Original description

NLLB - 200

This is the model card of NLLB - 200's distilled 1.3B variant. Here are the metrics for that particular checkpoint.

Intended Use

• Primary intended uses: NLLB - 200 is a machine translation model primarily intended for research in machine translation, especially for low - resource languages. It allows for single sentence translation among 200 languages. Information on how to use the model can be found in the Fairseq code repository along with the training code and references to evaluation and training data.
• Primary intended users: Primary users are researchers and the machine translation research community.
• Out - of - scope use cases: NLLB - 200 is a research model and is not released for production deployment. NLLB - 200 is trained on general domain text data and is not intended to be used with domain - specific texts, such as medical or legal domains. The model is not intended to be used for document translation. The model was trained with input lengths not exceeding 512 tokens, so translating longer sequences might result in quality degradation. NLLB - 200 translations cannot be used as certified translations.

Metrics

• Model performance measures: The NLLB - 200 model was evaluated using BLEU, spBLEU, and chrF++ metrics widely adopted by the machine translation community. Additionally, human evaluation was performed with the XSTS protocol, and the toxicity of the generated translations was measured.

Evaluation Data

• Datasets: The Flores - 200 dataset is described in Section 4.
• Motivation: Flores - 200 was used as it provides full evaluation coverage of the languages in NLLB - 200.
• Preprocessing: Sentence - split raw text data was preprocessed using SentencePiece. The SentencePiece model is released along with NLLB - 200.

Training Data

• Parallel multilingual data from various sources was used to train the model. A detailed report on data selection and construction process is provided in Section 5 of the paper. Monolingual data constructed from Common Crawl was also used, with more details in Section 5.2.

Ethical Considerations

• In this work, a reflexive approach was taken in technological development to prioritize human users and minimize risks. Many languages chosen for this study are low - resource languages, with an emphasis on African languages. While quality translation could improve education and information access, it could also make groups with lower digital literacy more vulnerable to misinformation or online scams. Regarding data acquisition, the training data was mined from various publicly available web sources. Although data cleaning was done, personally identifiable information may not be entirely eliminated. Finally, although efforts were made to optimize translation quality, mistranslations may still occur, which could have adverse impacts on decision - making, especially related to health and safety.

Caveats and Recommendations

• The model has been tested on the Wikimedia domain with limited investigation on other domains supported in NLLB - MD. Additionally, the supported languages may have variations that the model does not capture. Users should make appropriate assessments.

Carbon Footprint Details

• The carbon dioxide (CO2e) estimate is reported in Section 8.8.