"small language model" meaning in English

See small language model in All languages combined, or Wiktionary

Noun

Forms: small language models [plural]
Head templates: {{en-noun|head=small language model}} small language model (plural small language models)
  1. (machine learning) A type of neural network specializing in language but of a smaller scale, typically including millions of parameters, instead of billions to trillions. Wikidata QID: Q123759530 Synonyms: SLM Hypernyms: LM, language model, model Derived forms: SLM Related terms: large language model Coordinate_terms: LLM, large language model Translations (Translations): 小型语言模型 (Chinese Mandarin), მცირე ენობრივი მოდელი (mcire enobrivi modeli) (Georgian), mały model językowy [masculine] (Polish)

Inflected forms

Alternative forms

{
  "forms": [
    {
      "form": "small language models",
      "tags": [
        "plural"
      ]
    }
  ],
  "head_templates": [
    {
      "args": {
        "head": "small language model"
      },
      "expansion": "small language model (plural small language models)",
      "name": "en-noun"
    }
  ],
  "lang": "English",
  "lang_code": "en",
  "pos": "noun",
  "senses": [
    {
      "categories": [
        {
          "kind": "other",
          "name": "English entries with incorrect language header",
          "parents": [],
          "source": "w"
        },
        {
          "kind": "other",
          "name": "Entries with translation boxes",
          "parents": [],
          "source": "w"
        },
        {
          "kind": "other",
          "name": "Pages with 1 entry",
          "parents": [],
          "source": "w"
        },
        {
          "kind": "other",
          "name": "Pages with entries",
          "parents": [],
          "source": "w"
        },
        {
          "kind": "other",
          "name": "Terms with Georgian translations",
          "parents": [],
          "source": "w"
        },
        {
          "kind": "other",
          "name": "Terms with Mandarin translations",
          "parents": [],
          "source": "w"
        },
        {
          "kind": "other",
          "name": "Terms with Polish translations",
          "parents": [],
          "source": "w"
        },
        {
          "kind": "other",
          "langcode": "en",
          "name": "Linguistics",
          "orig": "en:Linguistics",
          "parents": [],
          "source": "w"
        },
        {
          "kind": "other",
          "langcode": "en",
          "name": "Machine learning",
          "orig": "en:Machine learning",
          "parents": [],
          "source": "w"
        }
      ],
      "coordinate_terms": [
        {
          "word": "LLM"
        },
        {
          "word": "large language model"
        }
      ],
      "derived": [
        {
          "word": "SLM"
        }
      ],
      "examples": [
        {
          "bold_text_offsets": [
            [
              493,
              514
            ]
          ],
          "ref": "2025 April 13, Stephen Ornes, “Small Language Models Are the New Rage, Researchers Say. Larger models can pull off a wider variety of feats, but the reduced footprint of smaller models makes them attractive tools”, in Wired:",
          "text": "Large language models work well because they’re so large. The latest models from OpenAI, Meta, and DeepSeek use hundreds of billions of “parameters” […] With more parameters, the models are better able to identify patterns and connections, which in turn makes them more powerful and accurate. But this power comes at a cost […] huge computational resources […] energy hogs […] In response, some researchers are now thinking small. IBM, Google, Microsoft, and OpenAI have all recently released small language models (SLMs) that use a few billion parameters—a fraction of their LLM counterparts. Small models are not used as general-purpose tools like their larger cousins. But they can excel on specific, more narrowly defined tasks, such as summarizing conversations, answering patient questions as a health care chatbot, and gathering data in smart devices. “For a lot of tasks, an 8 billion–parameter model is actually pretty good,” said Zico Kolter, a computer scientist at Carnegie Mellon University. They can also run on a laptop or cell phone, instead of a huge data center. (There’s no consensus on the exact definition of “small,” but the new models all max out around 10 billion parameters.) To optimize the training process for these small models, researchers use a few tricks. […]",
          "type": "quotation"
        }
      ],
      "glosses": [
        "A type of neural network specializing in language but of a smaller scale, typically including millions of parameters, instead of billions to trillions."
      ],
      "hypernyms": [
        {
          "word": "LM"
        },
        {
          "word": "language model"
        },
        {
          "word": "model"
        }
      ],
      "id": "en-small_language_model-en-noun-en:Q123759530",
      "links": [
        [
          "machine learning",
          "machine learning"
        ],
        [
          "neural network",
          "neural network"
        ],
        [
          "language",
          "language"
        ],
        [
          "million",
          "million"
        ],
        [
          "parameter",
          "parameter"
        ],
        [
          "billion",
          "billion"
        ],
        [
          "trillion",
          "trillion"
        ]
      ],
      "qualifier": "machine learning",
      "raw_glosses": [
        "(machine learning) A type of neural network specializing in language but of a smaller scale, typically including millions of parameters, instead of billions to trillions."
      ],
      "related": [
        {
          "word": "large language model"
        }
      ],
      "senseid": [
        "en:Q123759530"
      ],
      "synonyms": [
        {
          "word": "SLM"
        }
      ],
      "translations": [
        {
          "code": "cmn",
          "lang": "Chinese Mandarin",
          "lang_code": "cmn",
          "sense": "Translations",
          "word": "小型语言模型"
        },
        {
          "code": "ka",
          "lang": "Georgian",
          "lang_code": "ka",
          "roman": "mcire enobrivi modeli",
          "sense": "Translations",
          "word": "მცირე ენობრივი მოდელი"
        },
        {
          "code": "pl",
          "lang": "Polish",
          "lang_code": "pl",
          "sense": "Translations",
          "tags": [
            "masculine"
          ],
          "word": "mały model językowy"
        }
      ],
      "wikidata": [
        "Q123759530"
      ]
    }
  ],
  "word": "small language model"
}
{
  "derived": [
    {
      "word": "SLM"
    }
  ],
  "forms": [
    {
      "form": "small language models",
      "tags": [
        "plural"
      ]
    }
  ],
  "head_templates": [
    {
      "args": {
        "head": "small language model"
      },
      "expansion": "small language model (plural small language models)",
      "name": "en-noun"
    }
  ],
  "lang": "English",
  "lang_code": "en",
  "pos": "noun",
  "related": [
    {
      "word": "large language model"
    }
  ],
  "senses": [
    {
      "categories": [
        "English countable nouns",
        "English entries with incorrect language header",
        "English lemmas",
        "English multiword terms",
        "English nouns",
        "English terms with quotations",
        "Entries with translation boxes",
        "Pages with 1 entry",
        "Pages with entries",
        "Terms with Georgian translations",
        "Terms with Mandarin translations",
        "Terms with Polish translations",
        "Translation table header lacks gloss",
        "en:Linguistics",
        "en:Machine learning"
      ],
      "coordinate_terms": [
        {
          "word": "LLM"
        },
        {
          "word": "large language model"
        }
      ],
      "examples": [
        {
          "bold_text_offsets": [
            [
              493,
              514
            ]
          ],
          "ref": "2025 April 13, Stephen Ornes, “Small Language Models Are the New Rage, Researchers Say. Larger models can pull off a wider variety of feats, but the reduced footprint of smaller models makes them attractive tools”, in Wired:",
          "text": "Large language models work well because they’re so large. The latest models from OpenAI, Meta, and DeepSeek use hundreds of billions of “parameters” […] With more parameters, the models are better able to identify patterns and connections, which in turn makes them more powerful and accurate. But this power comes at a cost […] huge computational resources […] energy hogs […] In response, some researchers are now thinking small. IBM, Google, Microsoft, and OpenAI have all recently released small language models (SLMs) that use a few billion parameters—a fraction of their LLM counterparts. Small models are not used as general-purpose tools like their larger cousins. But they can excel on specific, more narrowly defined tasks, such as summarizing conversations, answering patient questions as a health care chatbot, and gathering data in smart devices. “For a lot of tasks, an 8 billion–parameter model is actually pretty good,” said Zico Kolter, a computer scientist at Carnegie Mellon University. They can also run on a laptop or cell phone, instead of a huge data center. (There’s no consensus on the exact definition of “small,” but the new models all max out around 10 billion parameters.) To optimize the training process for these small models, researchers use a few tricks. […]",
          "type": "quotation"
        }
      ],
      "glosses": [
        "A type of neural network specializing in language but of a smaller scale, typically including millions of parameters, instead of billions to trillions."
      ],
      "hypernyms": [
        {
          "word": "LM"
        },
        {
          "word": "language model"
        },
        {
          "word": "model"
        }
      ],
      "links": [
        [
          "machine learning",
          "machine learning"
        ],
        [
          "neural network",
          "neural network"
        ],
        [
          "language",
          "language"
        ],
        [
          "million",
          "million"
        ],
        [
          "parameter",
          "parameter"
        ],
        [
          "billion",
          "billion"
        ],
        [
          "trillion",
          "trillion"
        ]
      ],
      "qualifier": "machine learning",
      "raw_glosses": [
        "(machine learning) A type of neural network specializing in language but of a smaller scale, typically including millions of parameters, instead of billions to trillions."
      ],
      "senseid": [
        "en:Q123759530"
      ],
      "synonyms": [
        {
          "word": "SLM"
        }
      ],
      "wikidata": [
        "Q123759530"
      ]
    }
  ],
  "translations": [
    {
      "code": "cmn",
      "lang": "Chinese Mandarin",
      "lang_code": "cmn",
      "sense": "Translations",
      "word": "小型语言模型"
    },
    {
      "code": "ka",
      "lang": "Georgian",
      "lang_code": "ka",
      "roman": "mcire enobrivi modeli",
      "sense": "Translations",
      "word": "მცირე ენობრივი მოდელი"
    },
    {
      "code": "pl",
      "lang": "Polish",
      "lang_code": "pl",
      "sense": "Translations",
      "tags": [
        "masculine"
      ],
      "word": "mały model językowy"
    }
  ],
  "word": "small language model"
}

Download raw JSONL data for small language model meaning in English (3.7kB)


This page is a part of the kaikki.org machine-readable English dictionary. This dictionary is based on structured data extracted on 2026-03-25 from the enwiktionary dump dated 2026-03-03 using wiktextract (05c257f and 9d9a410). The data shown on this site has been post-processed and various details (e.g., extra categories) removed, some information disambiguated, and additional data merged from other sources. See the raw data download page for the unprocessed wiktextract data.

If you use this data in academic research, please cite Tatu Ylonen: Wiktextract: Wiktionary as Machine-Readable Structured Data, Proceedings of the 13th Conference on Language Resources and Evaluation (LREC), pp. 1317-1325, Marseille, 20-25 June 2022. Linking to the relevant page(s) under https://kaikki.org would also be greatly appreciated.