OPT
E435873
OPT is a family of open-source large language models developed by Meta AI, designed as efficient, GPT-style transformer models for natural language processing tasks.
All labels observed (1)
| Label | Occurrences |
|---|---|
| OPT canonical | 1 |
How this entity was disambiguated
This entity first appeared as the object of triple T4389200 — resolving that mention is where its identity was fixed. The disambiguator weighed these candidate entities and picked the highlighted one (or “None”, minting a new entity). This is how homonymy is resolved: the same surface form can point to different entities.
NED1
Entity disambiguation (via context triple)
gpt-5-mini-2025-08-07
Target entity: OPT Context triple: [Hugging Face Transformers, supportsModelType, OPT]
-
A.
optio
An optio was a junior officer in the Roman army who served as the deputy and second-in-command to a centurion within a legionary unit.
-
B.
OPS
OPS is the abbreviated name for the Office of Peace Operations, Sanctions, and Counterterrorism, a U.S. State Department bureau focused on peacekeeping, sanctions policy, and counterterrorism efforts.
-
C.
Opti
Opti is a friendly, futuristic robot character that served as one of the official mascots of Expo 2020 Dubai.
-
D.
OPP
OPP is a division within the Technology Transformation Services focused on managing and delivering digital products and programs across the U.S. federal government.
-
E.
the O
The O is MONA’s custom digital guide app that provides visitors with interactive information, commentary, and navigation throughout the Museum of Old and New Art.
- F. None of above. chosen
- G. Unsure - the case is ambiguous/there is not enough information to decide.
NED2
Entity disambiguation (via description)
gpt-5-mini-2025-08-07
Target entity: OPT Target entity description: OPT is a family of open-source large language models developed by Meta AI, designed as efficient, GPT-style transformer models for natural language processing tasks.
-
A.
optio
An optio was a junior officer in the Roman army who served as the deputy and second-in-command to a centurion within a legionary unit.
-
B.
OPS
OPS is the abbreviated name for the Office of Peace Operations, Sanctions, and Counterterrorism, a U.S. State Department bureau focused on peacekeeping, sanctions policy, and counterterrorism efforts.
-
C.
Opti
Opti is a friendly, futuristic robot character that served as one of the official mascots of Expo 2020 Dubai.
-
D.
OPP
OPP is a division within the Technology Transformation Services focused on managing and delivering digital products and programs across the U.S. federal government.
-
E.
the O
The O is MONA’s custom digital guide app that provides visitors with interactive information, commentary, and navigation throughout the Museum of Old and New Art.
- F. None of above. chosen
Statements (49)
| Predicate | Object |
|---|---|
| instanceOf |
autoregressive language model
ⓘ
large language model family ⓘ |
| basedOn | GPT-style architecture ⓘ |
| codeRepositoryHost | GitHub ⓘ |
| codeRepositoryOrganization |
facebookresearch
ⓘ
linked to:
Meta AI
|
| comparisonTarget |
OpenAI GPT-3
ⓘ
linked to:
GPT-3
|
| designGoal |
efficient GPT-style model
ⓘ
open-source alternative to GPT-3 ⓘ |
| developer |
Meta AI
ⓘ
Meta Platforms ⓘ
linked to:
Meta Platforms, Inc.
|
| hasPaper | OPT: Open Pre-trained Transformer Language Models ⓘ |
| includesVariant |
OPT-1.3B
ⓘ
OPT-125M ⓘ OPT-13B ⓘ OPT-175B ⓘ OPT-2.7B ⓘ OPT-30B ⓘ OPT-350M ⓘ OPT-6.7B ⓘ OPT-66B ⓘ |
| intendedUse |
benchmarking against GPT-3
ⓘ
downstream NLP applications ⓘ research ⓘ |
| language | English ⓘ |
| largestModel | OPT-175B ⓘ |
| license | custom Meta license ⓘ |
| modelArchitecture | transformer ⓘ |
| modelType | decoder-only transformer ⓘ |
| openSource | true ⓘ |
| organization | Meta AI ⓘ |
| paperArchive | arXiv ⓘ |
| paperArxivId | 2205.01068 ⓘ |
| parameterRange | 125M–175B parameters ⓘ |
| releaseDate | 2022-05 ⓘ |
| supportsFineTuning | true ⓘ |
| supportsInferencePrecision |
BF16
ⓘ
FP16 ⓘ |
| supportsPrompting | true ⓘ |
| supportsTask |
classification via prompting
ⓘ
dialogue modeling ⓘ language modeling ⓘ question answering ⓘ summarization ⓘ text completion ⓘ text generation ⓘ |
| trainingComputeOptimization | efficiency-focused implementation ⓘ |
| trainingDataSource |
licensed text data
ⓘ
publicly available text data ⓘ |
| trainingObjective | causal language modeling ⓘ |
How these facts were elicited
The pipeline generated the facts above by prompting gpt-5.1 with this entity's name + description and the instruction below.
Instruction
You are a knowledge base construction expert. Given a subject entity and a description of it, return factual statements that you know for the subject as a JSON list of dictionaries(triples), where keys must be "subject", "predicate" and "object". The number of facts may be very high, between 25 to 50 or more, for very popular subjects. For less popular subjects, the number of facts can be very low, like 5 or 10. # Requirements - If you don't know the subject at all, return an empty list. - If the subject is not a named entity, return an empty list. - Include at least one triple where predicate is "instanceOf". - Do not get too wordy. - Separate several objects into multiple triples with one object.
Input
Subject: OPT Description of subject: OPT is a family of open-source large language models developed by Meta AI, designed as efficient, GPT-style transformer models for natural language processing tasks.
Referenced by (1)
Full triples — surface form annotated when it differs from this entity's canonical label.