# microsoft/DialoGPT

Large-scale pretraining for dialogue

Repository: https://github.com/microsoft/DialoGPT
Canonical: https://ross.abutalabs.com/products/dialogpt
Language: Python
License: MIT
License Family: permissive
Topics: dialogue, machine-learning, pytorch, transformer, text-generation, dialogpt, gpt-2, text-data, data-processing
Archived: true
Last push: 2022-10-17T23:41:52+00:00

## Health v2 (maintenance only)
Score: 10/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 0, release rhythm 35, longevity 100
- inputs: {"age_days": 2561, "days_push": 1416, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: no_releases, archived
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 2420, forks 349 (observed 2026-08-28T04:06:50.521267+00:00)

## What it is
DialoGPT is a large-scale pretrained dialogue response generation model from Microsoft, based on GPT-2 and trained on 147M multi-turn Reddit conversations. The repository provides training code, data extraction scripts, and pretrained small/medium/large model checkpoints, though the project is now superseded by GODEL.

## Use cases
- generate conversational responses with a pretrained chatbot model
- fine-tune a dialogue model on my own conversation data
- reproduce DialoGPT results for a research paper
- build a single-turn chatbot in PyTorch
- download pretrained dialogue model checkpoints
- extract and process Reddit dialogue data for training

## When to choose
- you need reproducibility of the original DialoGPT paper results
- you want a lightweight GPT-2-based chatbot baseline for research
- you need pretrained dialogue checkpoints in 117M/345M/762M sizes

## When to avoid
- you want the best dialogue quality - use GODEL or a modern LLM instead
- you need production-grade multi-turn conversation with safety features
- you want an actively maintained project

## Facets
- artifact type: library
- maturity: maintenance
- function: llm-training, machine-learning, chatbot, nlp
- domain: large-language-models, chatbots, machine-learning
- platform: python
- tags: dialogue-generation, gpt-2, pytorch, pretrained-model, response-generation, reddit-data, superseded-by-godel, natural-language-processing, gpu, linux

## Member repositories
- microsoft/DialoGPT (main) score 10

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:06:50.521267+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-30T02:33:05.175800+00:00, confidence not recorded.
  - readme: https://github.com/microsoft/DialoGPT (fetched 2026-08-28T04:06:50.521267+00:00, sha 6dad731ae81a)
- Data as of 2026-08-30T08:39:29.467469+00:00.