erwanf / text-generation Open weights updated 2 years ago

gpt2-mini

Name: gpt2-mini
Author: erwanf

A smaller GPT-2 model with (only) 39M parameters. It was pretrained on a subset of OpenWebText, the open-source version of the pretraining dataset used by OpenAI for the original GPT-2 models.

Params

39 M

Context

—

Downloads 30d

285 K

Likes

Commercial use: allowed mit Not gated SAFETENSORS 1 languages View on Hugging Face ↗

Download history

daily snapshots · 15 days

285 K137 K

Oct 27Nov 1Nov 6Nov 10

Can you run it?

Estimated VRAM at 8K context unless noted. Pick your hardware to see the verdict per quantization.

File	Quant	Size	Est. VRAM	Verdict on RTX 4090 · 24 GB
model.safetensors	f32	0.2 GB	0.7 GB	✅ Runs comfortably

Estimate: file size × 1.1 + KV cache at 8K + 0.5 GB overhead. Not a benchmark — how we calculate this.

Run it

copy-paste, exact tags checked against the Hub

~ · curl · api/v1

$ curl -s https://aimodelscomparison.com/api/v1/models/gpt2-mini
{
  "hf_id": "erwanf/gpt2-mini",
  "params_b": 0.04,
  "context_length": null,
  "license": { "id": "mit", "commercial": "yes" },
  "downloads_30d": 285168,
  "vram_estimates": [
    { "quant": "f32", "gb": 0.7 }
  ],
  "updated_at": "2026-10-27T01:00:26Z"
}

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

Specifications

Architecture: GPT2LMHeadModel
Parameters: 39 M
Tensor type: F32
Vocabulary: 50,257
Licence: mit
First seen on the Hub: 2024-06-23
Training datasets: Skylion007/openwebtext
Added to our catalog: 2026-10-27

Compare with any text-generation model

Popular comparisons

gpt2 31.8 M opt-125m 4.7 M SmolLM2-135M 2.5 M distilgpt2 1.5 M SmolLM2-135M-Instruct 847 K pythia-70m-deduped 729 K

Appears in

Trending text-generation