DataLearner logo
PA

PaLM

Foundation modelPaLM

Pathways Language Model

Release date: 2022-04-05Updated: 2023-08-16Views: 809
Live demoGitHubHugging FaceCompare
Parameters
540B
Context length
2K
Multilingual
No data
Reasoning ability
No data

Pathways Language Model is an AI model published by Google Research, released on 2022-04-05, for Foundation model, with 540B parameters, and 2K context length, with a 88.40 score on HumanEval.

Data sourced primarily from official releases (GitHub, Hugging Face, papers), then benchmark leaderboards, then third-party evaluators. Learn about our data methodology

PaLM

Model basics

Reasoning traces
No data
Thinking modes
Thinking modes not supported
Context length
2K tokens
Max output length
No data
Model type
Foundation model
Modality (in / out)
No data
Release date
2022-04-05
Model file size
No data
MoE architecture
No
Total params / Active params
540B / Not applicable
Knowledge cutoff
No data
PaLM

Open source & experience

Code license
Proprietary
Weights license
Proprietary
GitHub repo
N/A
Hugging Face
N/A
Live demo
N/A
PaLM

Official resources

Paper
DataLearnerAI blog
N/A
PaLM

API details

API speed
No data
No public API pricing yet.
PaLM

Benchmark Results

PaLM currently shows benchmark results led by HumanEval (23 / 140, score 88.40), GSM8K (40 / 70, score 56.50), MMLU (76 / 124, score 69.30). This page also consolidates core specs, context limits, and API pricing so you can evaluate the model from benchmark results and deployment constraints together.

Thinking

General Knowledge

1 evaluations
Benchmark / mode
Score
Rank/total
MMLU
Standard Mode
69.30
76 / 124

Math and Reasoning

1 evaluations
Benchmark / mode
Score
Rank/total
GSM8K
Standard Mode
56.50
40 / 70

Coding and Software Engineer

3 evaluations
Benchmark / mode
Score
Rank/total
HumanEval
Standard Mode
88.40
23 / 140
HumanEval
Standard Mode
26.20
125 / 140
MBPP
Standard Mode
47
75 / 96
PaLM

Publisher

Pathways Language Model

Model Overview

Pathways Language Model is an AI model published by Google Research, released on 2022-04-05, for Foundation model, with 540B parameters, and 2K context length, with a 88.40 score on HumanEval.

DataLearner on WeChat

Follow DataLearner on WeChat for AI model updates and research notes.

DataLearner WeChat QR code