DataLearner logo
PY

Pythia

Foundation model

Pythia

Release date: 2023-04-03Updated: 2023-08-16 22:08:35.518767
Parameters
12B
Context length
2K
Chinese support
Not supported
Reasoning ability

Pythia is an AI model published by EleutherAI, released on 2023-04-03, for Foundation model, with 12B parameters, and 2K context length, requiring about 24GB storage, with a 26.20 score on C-Eval.

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

Pythia

Model basics

Reasoning traces
Not supported
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
2023-04-03
Model file size
24GB
MoE architecture
No
Total params / Active params
12B / N/A
Knowledge cutoff
No data
Pythia

Open source & experience

Code license
Weights license
Apache 2.0- Commercial use permitted
Live demo
No live demo
Pythia

Official resources

Paper
DataLearnerAI blog
No blog post yet
Pythia

API details

API speed
No data
No public API pricing yet.
Pythia

Benchmark Results

Pythia currently shows benchmark results led by AGIEval (25 / 28, score 25.30), C-Eval (44 / 48, score 26.20), MMLU (123 / 124, score 25.10). 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

2 evaluations
Benchmark / mode
Score
Rank/total
C-Eval
Standard Mode
26.20
44 / 48
MMLU
Standard Mode
25.10
123 / 124

General Evaluation

1 evaluations
Benchmark / mode
Score
Rank/total
AGIEval
Standard Mode
25.30
25 / 28

Compare with other models

No curated comparisons for this model yet.

Want a custom combination? Open the compare tool

Pythia

Publisher

Pythia

Model Overview

Pythia is an AI model published by EleutherAI, released on 2023-04-03, for Foundation model, with 12B parameters, and 2K context length, requiring about 24GB storage, with a 26.20 score on C-Eval.

DataLearner on WeChat

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

DataLearner WeChat QR code