DataLearner logo
MO

MOSS

Foundation model

MOSS

Release date: 2023-02-20Updated: 2023-07-10Views: 795
Live demoGitHubHugging FaceCompare
Parameters
16B
Context length
2K
Multilingual
Supported
Reasoning ability
No data

MOSS is an AI model published by OpenLMLab, released on 2023-02-20, for Foundation model, with 16B parameters, and 2K context length, with a 33.13 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

MOSS

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
2023-02-20
Model file size
No data
MoE architecture
No
Total params / Active params
16B / Not applicable
Knowledge cutoff
No data
MOSS

Open source & experience

Code license
Weights license
- Commercial use permitted
Hugging Face
N/A
Live demo
N/A
MOSS

Official resources

Paper
N/A
DataLearnerAI blog
N/A
MOSS

API details

API speed
No data
No public API pricing yet.
MOSS

Benchmark Results

MOSS currently shows benchmark results led by AGIEval (23 / 28, score 26.80), C-Eval (41 / 48, score 33.13), MMLU (121 / 124, score 27.40). 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
33.13
41 / 48
MMLU
Standard Mode
27.40
121 / 124

General Evaluation

1 evaluations
Benchmark / mode
Score
Rank/total
AGIEval
Standard Mode
26.80
23 / 28
MOSS

Publisher

MOSS

Model Overview

MOSS is an AI model published by OpenLMLab, released on 2023-02-20, for Foundation model, with 16B parameters, and 2K context length, with a 33.13 score on C-Eval.

DataLearner on WeChat

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

DataLearner WeChat QR code