WeLM

Closed weights WeChat AI 10B parameters May 2023

No estimate

No hardware requirements for this model

The weights for this model have not been published, so it cannot be downloaded or run on your own hardware at any size. It is reachable only through its provider, and no graphics card changes that.

On record

Full specification

Everything on record for this model. Most of it describes how it was trained rather than how it runs — useful context for judging how much work went into it, and how it compares with models built at a different scale.

Origin

Who built this model, where, and when it was published.

Organisation
WeChat AI
Organisation type
Industry
Country
China
Published
16 May 2023
Authors
Hui Su, Xiao Zhou, Houjin Yu, Xiaoyu Shen, Yuwen Chen, Zilin Zhu, Yang Yu, Jie Zhou

What it does

The problem areas the model was built for. A model can carry several of each.

Domain
Language
Task
Language modeling/generation, Translation
Approach
Unsupervised

Size

How large the model is and how much data it was trained on. Parameters are the figure that decides whether it fits on a given graphics card.

Parameters
10B

"WeLM is trained with 10B parameters”

Training data
262,000,000,000 tokens

from the paper "After all the above filtering process, our corpus contains 262B tokens”

Training compute

The arithmetic performed to train the model, measured in floating-point operations. It is a measure of what the training run cost, not of how fast the finished model answers you.

Training compute
2.5 × 10²² FLOP

"The largest model is trained on 128 A100-SXM4-40GB GPUs in about 24 days”, "All models are trained with FP16 mixed precision." 3.12e14 * 128 * 24 * 24 * 3600 * 0.3 = 2.48e22 (FLOP/GPU-s) * (GPU) * (days) * (h/day) * (sec/h) * (utilization assumption)

How it was established
Hardware

The training run

What it physically took to train: which chips, how many, for how long, and what that drew from the wall.

Training hardware
NVIDIA A100 SXM4 40 GB
Chips used
128
Chip-hours
73,728
Wall-clock time
576 hours (24 days)

"The largest model is trained on 128 A100-SXM4-40GB GPUs in about 24 days”,

Power draw
102.0 kW

How it is classified

Labels the source dataset applies when tracking notable models, and how confident it is in the entry.

Record confidence
Confident
Citations
25

Sources

Where this record came from and when it was last checked.

Reference
WeLM: A Well-Read Pre-trained Language Model for Chinese
Last updated
25 May 2026

What the numbers mean

Where it came from

WeLM was published by WeChat AI, in China, in May 2023. It comes out of industry.

It works in Language, and is recorded as doing language modeling/generation, Translation.

This is a closed model: the trained values stayed with whoever produced them, and there is no local version to run.

Training and provenance

The training run consumed about 2.5 × 10²² FLOP, on NVIDIA A100 SXM4 40 GB. That figure describes the cost of creating it and has no bearing on how quickly it generates text.

The training set ran to roughly 262,000,000,000 tokens.

Answers

WeLM — common questions

01

What is WeLM used for?

WeLM works in Language, and is recorded as handling language modeling/generation, Translation. A model can carry several of each, so these are the areas it was built for rather than a limit on what it will attempt.

02

How much compute was used to train WeLM?

Around 2.5 × 10²² FLOP, on NVIDIA A100 SXM4 40 GB. That measures what producing the model cost and says nothing about how quickly it answers once trained — inference speed comes from memory bandwidth, not from the training budget.

03

What GPU do I need to run WeLM?

None. WeLM is a closed model — its weights were never published, so it cannot be downloaded or run on your own hardware at any price. It is reachable only through its provider.

04

Is WeLM open source?

The licensing for WeLM was never recorded in our source data. We treat unstated licensing as closed, because an unrecorded licence is not one to rely on.

05

How many parameters does WeLM have?

WeLM has 10B parameters. "WeLM is trained with 10B parameters”. That figure is the total, and it is what decides how much memory the model needs — roughly half a gigabyte per billion at the compression most people use.

06

Who created WeLM?

WeLM was published by WeChat AI, based in China, categorised as industry.

07

When was WeLM released?

WeLM was published in May 2023. Capability per parameter has improved considerably since, so a newer model of the same size is often the better use of the same hardware.

Source

Original publication

Record last updated 25 May 2026

The other direction

Looking at it from the other side?

This page starts from the model. If you already own a card and want to know everything it will run, start from the hardware instead.