Name: Opt 1.3B
Author: facebook

Opt 1.3B is an open-source language model by facebook. Features: 1.3b LLM, VRAM: 2.6GB, Context: 2K, License: other, HF Score: 34.6, LLM Explorer Score: 0.15, Arc: 29.5, HellaSwag: 54.5, MMLU: 25, TruthfulQA: 38.7, WinoGrande: 59.8, GSM8K: 0.2.

Arxiv:2005.14165 Arxiv:2205.01068 Autotrain compatible En Jax Opt Pytorch Region:us Tf

Model Card on HF 🤗: https://huggingface.co/facebook/opt-1.3b

Opt 1.3B Benchmarks

MMLU Pro: 1.19

MUSR: 2.08

BBH: 3.65

IFEval: 23.83 vs 88 (so35)^-72.9%

ARC: 29.52 vs 96.7 (so35)^-69.5%

HellaSwag: 54.53 vs 95.3 (gpt4)^-42.8%

MMLU: 24.96 vs 88.3 (so35)^-71.7%

TruthfulQA: 38.71 vs 59 (gpt4)^-34.4%

WinoGrande: 59.75 vs 87.5 (gpt4)^-31.7%

GSM8K: 0.15 vs 96.4 (so35)^-99.8%

MATH Lvl 5: 0.91

LLME Score: 0.14669

^nn.n% — How the model compares to the reference models: Anthropic Sonnet 3.5 ("so35"), GPT-4o ("gpt4o") or GPT-4 ("gpt4").

What is the LLM Explorer Rank (Score)

🌟 Advertise your project 🚀

Opt 1.3B Parameters and Internals

Model Type

decoder-only, causal language model

Use Cases

Areas:

research, evaluation of downstream tasks, text generation

Primary Use Cases:

text generation, evaluation of downstream tasks, fine-tuning

Limitations:

The model has biases due to unfiltered web data, Quality issues in generation diversity and hallucination

Supported Languages

English (Predominantly), Non-English (Minimal)

Training Details

Data Sources:

BookCorpus, CC-Stories, The Pile, Pushshift.io Reddit dataset, CCNewsV2

Data Volume:

180B tokens

Methodology:

Causal language modeling

Context Length:

2048

Training Time:

~33 days

Hardware Used:

992 80GB A100 GPUs

Model Architecture:

Decoder-only pre-trained transformer

Input Output

Input Format:

Sequences of 2048 tokens

Accepted Modalities:

text

Output Format:

Text generation

LLM Name	Opt 1.3B
Repository 🤗	https://huggingface.co/facebook/opt-1.3b
Model Size	1.3b
Required VRAM	2.6 GB
Updated	2025-09-23
Maintainer	facebook
Model Type	opt
Model Files	2.6 GB
Supported Languages	en
Model Architecture	OPTForCausalLM
License	other
Context Length	2048
Model Max Length	2048
Transformers Version	4.21.0.dev0
Beginning of Sentence Token	</s>
End of Sentence Token	</s>
Unk Token	</s>
Vocabulary Size	50272
Torch Data Type	float16
Activation Function	relu
Errors	replace

Quantized Models of the Opt 1.3B

Model	Likes	Downloads	VRAM
Opt 1.3B GPTQ 4bit G128	0	5	1 GB
Opt 1.3B Gptq 4bit	0	7	1 GB

Best Alternatives to Opt 1.3B

Best Alternatives	Context / RAM	Downloads	Likes
LLmRa 1.3B V2	2K / 5.3 GB	664	0
Sparse0.5 OPT 1.3	2K / 2.6 GB	1821	0
... Distilled PromptKD Dolly V1.0	2K / 2.6 GB	6	0
New Galactica 1.3b Mcq	2K / 2.6 GB	5	0
... Galactica 1.3b DPO 0.1beta Ai	2K / 2.6 GB	5	1
New Galactica 1.3b Mcq No Rag	2K / 2.6 GB	6	0
New Galactica 1.3b Mcq Rag	2K / 2.6 GB	6	0
... Galactica 1.3b DPO 0.1beta Ai	2K / 2.6 GB	5	0
...tica 1.3b 0.1beta Ai Raft Top3	2K / 2.6 GB	5	0
...tica 1.3b 0.1beta Ai Raft Top4	2K / 2.6 GB	5	0

Note: green Score (e.g. "73.2") means that the model is better than facebook/opt-1.3b.

Rank the Opt 1.3B Capabilities

🆘 Have you tried this model? Rate its performance. This feedback would greatly assist ML community in identifying the most suitable model for their needs. Your contribution really does make a difference! 🌟

Instruction Following and Task Automation
Factuality and Completeness of Knowledge
Censorship and Alignment
Data Analysis and Insight Generation
Text Generation
Text Summarization and Feature Extraction
Code Generation
Multi-Language Support and Translation

What open-source LLMs or SLMs are you in search of? 52052 in total.

Email us: info@extractum.io. Our Privacy Policy | Terms and Conditions | Suggest an improvement.

Our Social Media →

Original data from HuggingFace, OpenCompass and various public git repos.

Check out Ag3ntum — our secure, self-hosted AI agent for server management.

Release v20260327b

Support LLM Explorer

Opt 1.3B by facebook

» All LLMs » facebook » Opt 1.3B URL Share it on