Text

OPT

An early open Meta series matching GPT-3 in size. Outdated; useful for research and comparison.

The last open version came out in Dec 2022. The family has not been updated for a long time: the model still works, but do not expect fixes or new sizes.

Developer
Meta, USA
First release
May 2022
Latest release
Dec 2022
Sizes
125M – 66B (175B on request)
License
Non-commercial onlyOPT License: non-commercial and research use only
Russian
Not supported
Running
On your own serverAlso runs without a GPU
Industries
Science and research, Education
Deploy on your server Hugging Face GitHub

What it does

  • Research experiments
  • Training specialists
  • Comparison with modern models

Where it is used

Research groupsUniversities

Hardware requirements

LaptopLaptop or regular PC, up to 8 GB of VRAM — smaller versions
fits
1 GPUOne GPU with 16–80 GB — mid-size versions
fits
ClusterServer with several GPUs — flagship versions
fits

Versions

  1. OPT-IML
  2. OPT

How to run it

On your own serverWeights are downloaded from Hugging Face and served with vLLM (load and API) or llama.cpp (modest hardware). The model then runs in a closed network with no per-request fees.Hugging Face
How much hardware you needCalculate the VRAM for the model size, context length and number of concurrent requests.Open the hardware calculator

I can set this up end to end: pick the model size, deploy it on your server and connect it to your systems.

Frequently asked questions

Can OPT be used in a commercial project?

No. License: OPT License: non-commercial and research use only. A commercial product needs a different model or a separate agreement with the rights holder.

What hardware does OPT need?

At minimum: Laptop or regular PC, up to 8 GB of VRAM — smaller versions. Some versions also run on an ordinary CPU, without a GPU. You can calculate the exact VRAM for your model size and context in the hardware calculator.

Does OPT support Russian?

No. The model card lists its languages and Russian is not among them.

Where can I download OPT and what does it cost?

The OPT weights are open and free to download. You only pay for the hardware it runs on and for the setup. Source links are at the bottom of this page.

How I deploy it for clients

  1. SelectionI pick the model size for your task and hardware and test it on your examples.
  2. DeploymentI deploy it on your server or in a closed network and provide an API.
  3. Fine-tuningI fine-tune it on your data (LoRA) or connect a knowledge base — whichever is cheaper for the task.
  4. IntegrationI connect it to your CRM, ERP, bot, website or team chat and set up monitoring.

Similar models

TextPythia и GPT-NeoXEleutherAI · USACommercial use allowed

Fully open models from the non-profit lab EleutherAI: GPT-NeoX-20B and the Pythia series with published intermediate training checkpoints.

DetailsTextBLOOMBigScience (Hugging Face and the community) · FranceCommercial use with conditions

One of the first large open models, trained by a community of hundreds of researchers in 46 languages. Today it is interesting mostly as a historical milestone.

DetailsTextLlamaMeta · USACommercial use with conditions

The models that started mass open source in AI. A huge ecosystem of fine-tuned versions and tools.

Details

Source: huggingface.co/facebook/opt-66b. Data checked against the model card on 22 Sep 2026. Have a lawyer review the license before commercial launch.