ChatMol
No estimate
No hardware requirements for this model
This model's weights are open, but no parameter count has been published for it. Every memory and speed figure starts from that number, so we would rather show nothing than a fabricated estimate.
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
- Tsinghua University,PingAn Technology,Beijing University of Posts and Telecommunications
- Organisation type
- Academia,Industry,Academia
- Country
- China
- Published
- 2 September 2024
- Authors
- Zheni Zeng, Bangchen Yin, Shipeng Wang, Jiarui Liu, Cheng Yang, Haishen Yao, Xingzhi Sun, Maosong Sun, Guotong Xie, Zhiyuan Liu
What it does
The problem areas the model was built for. A model can carry several of each.
- Domain
- Biology
- Task
- Drug discovery
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.
- Training data
- tokens
Summary of calculations: ChEBI-20: 26,407 samples × 143 tokens/sample = 3,775,701 tokens PCdes: 10,500 samples × 162 tokens/sample = 1,701,000 tokens ChEBI-dia: 7,361 dialogs × 200 tokens/dialog = 1,472,200 tokens Total dataset tokens: 3,775,701 + 1,701,000 + 1,472,200 = 7,948,901 tokens Training tokens: 12,000 steps × 256 tokens/step = 3,072,000 tokens Final data estimate: 3.072 × 10^6 tokens
Availability
Whether you can obtain the model and run it on your own hardware, which is what decides if any of the graphics-card figures on this page apply.
- Weights
- Open — downloadable
- Model access
- Open weights (non-commercial)
- Training code
- Open (non-commercial)
Codes and data are provided in https://github.com/Ellenzzn/ChatMol/tree/main [no clear license]
How it is classified
Labels the source dataset applies when tracking notable models, and how confident it is in the entry.
- Record confidence
- Speculative
- Citations
- 13
Sources
Where this record came from and when it was last checked.
- Reference
- ChatMol: interactive molecular discovery with natural language
- Last updated
- 28 November 2025
What the numbers mean
About this model
ChatMol was published by Tsinghua University,PingAn Technology,Beijing University of Posts and Telecommunications, in China, in September 2024. It comes out of academia,Industry,Academia.
It works in Biology, and is recorded as doing drug discovery.
The weights are published, so it can be downloaded and run on your own hardware indefinitely, offline, with no account attached.
Answers
ChatMol — common questions
What is ChatMol used for?
ChatMol works in Biology, and is recorded as handling drug discovery. 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.
Where can I download ChatMol?
The weights for ChatMol are published, though we do not hold a repository link for it. This site calculates hardware requirements rather than hosting model files.
What GPU do I need to run ChatMol?
We cannot say. ChatMol has open weights, but no parameter count has been published for it, and every memory and speed calculation starts from that number. We would rather show nothing than a fabricated estimate.
Is ChatMol open source?
Its weights are published, so ChatMol can be downloaded and run on your own hardware. Note that open weights is not the same as open source in the full sense — it says nothing about the training data, the training code, or the commercial terms attached.
How many parameters does ChatMol have?
No parameter count has been published for ChatMol, which is why no memory or speed figure appears on this page.
Who created ChatMol?
ChatMol was published by Tsinghua University,PingAn Technology,Beijing University of Posts and Telecommunications, based in China, categorised as academia,Industry,Academia.
When was ChatMol released?
ChatMol was published in September 2024. Capability per parameter has improved considerably since, so a newer model of the same size is often the better use of the same hardware.
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.