facebook / text-generation updated 3 years ago

opt-125m

Name: opt-125m
Author: facebook

OPT was first introduced in Open Pre-trained Transformer Language Models and first released in metaseq's repository on May 3rd 2022 by Meta AI.

Params

125 M

Context

2,048

Downloads 30d

7.4 M

Likes

323

Licence: other Not gated 1 languages View on Hugging Face ↗

Download history

daily snapshots · 55 days

▲ 8.9 M in the last 30 days (54.7%)

18.6 M7.4 M

Jul 28Aug 15Sep 2Sep 20

Run it

copy-paste, exact tags checked against the Hub

~ · curl · api/v1

$ curl -s https://aimodelscomparison.com/api/v1/models/opt-125m
{
  "hf_id": "facebook/opt-125m",
  "params_b": 0.13,
  "context_length": 2048,
  "license": { "id": "other", "commercial": "unknown" },
  "downloads_30d": 7387919,
  "vram_estimates": [

  ],
  "updated_at": "2026-09-18T01:00:03Z"
}

est. VRAM —on RTX 4090 · 24 GBJSON API →

Specifications

Architecture: OPTForCausalLM
Parameters: 125 M
Context length: 2,048
Vocabulary: 50,272
Layers / heads: 12 / 12
Licence: other
First seen on the Hub: 2022-05-11
Training datasets: undisclosed
Added to our catalog: 2026-07-28

Compare with any text-generation model

Popular comparisons

gpt2 15.4 M Qwen2.5-0.5B-Instruct 8.5 M pythia-160m 3.5 M SmolLM2-135M 2.1 M distilgpt2 2.1 M SmolLM2-135M-Instruct 1.5 M

Appears in

Models that fit in 12 GB