Google unveils LaMDA

Made by AIWritten by AI (Claude) from the source below. Check anything important against the source.How it's made

LaMDA (Language Model for Dialogue Applications) unveiled — a Transformer-based conversation model trained on dialogue

Why it matters

Unlike most language models it was trained on dialogue, and Google said fine-tuning after training could significantly improve the sensibleness and specificity of its responses

Source — Primary sources first; secondary reports are labelled “Press”

Officialblog.googlehttps://blog.google/technology/ai/lamda/

Added in Week 41, 2026 · See that week

On the map — ⑫ Training Data line

  1. Previous stopThe Pile
  2. This record · stopLaMDA
  3. Next stopSelf-Instruct
See it on the map

In the model family tree — Gemini family · 21 versions

  1. Earlier versionT5
  2. This recordLaMDA
  3. Later versionPaLM
See the whole Gemini family

Related notes — Notes on the same axis from the same week

Post-training that sorts wrong answers into three conditions and treats each

Nearby on the same axis

The PilePaperGoogle releases the T5 paper and C4PaperSelf-InstructPaperRedPajama-Data-1TOfficial