Model

LaMDA

CLOSED

Google · LaMDA family · released May 18, 2021

137B dialogue-specialized decoder-only Transformer; predecessor behind Google's original Bard chatbot.

ReasoningCodingVisionFunction callingTool useAgentic

-180.4

Elo · rank #420

Parameters

137B

Active params

137B (dense)

Context

2K tokens

Architecture

Dense decoder-only Transformer (137B, dialogue-tuned)

License

Proprietary (Google; research model, weights never released)

Languages

—

API price (in/out)

No hosted API

Modalities

text

Benchmark results

Bar shows position within the tracked field; marker = field best

ANLIReasoning39.9%#7

best: Phi-3-small (7B) · 58.1%

ARC-EasyReasoning76.4%#34

best: Phi-3-medium (14B) · 97.7%

GSM8KMath14.3%#173

best: Llama 3.1 405B · 96.8%

HumanEvalCoding14.0%#179

best: Claude Opus 4.5 · 99.4%

MBPPCoding14.8%#108

best: Llama-3.3-Nemotron-Super-49B v1 (Reasoning On) · 91.3%

Run it locally

Closed weights — available via API only. No local deployment.

Input / M tok

—

Output / M tok

—

LaMDA family

Elo progression across releases

Resources

API price — · each benchmark row carries its own source badge (see methodology)