DeepSA
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
- ShanghaiTech University
- Organisation type
- Academia
- Country
- China
- Published
- 2 November 2023
- Authors
- Shihang Wang, Lin Wang, Fenglei Li, Fang Bai
What it does
The problem areas the model was built for. A model can carry several of each.
- Domain
- Biology
- Task
- Synthetic accessibility model
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
- Epochs
- 20
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
- 1 × 10¹⁹ FLOP
- How it was established
- Hardware
1. Hardware setup: 1x NVIDIA GeForce RTX 3090 (1.60 × 10¹⁴ FLOP/s) 2. Training duration: Estimated based on dataset size and batch processing - 71,861,060 total samples (3,593,053 samples × 20 epochs) - 561,719 batches (batch size 128) - ~15.6 hours (56,172 seconds) 3. Utilization rate: 40% 4. Final calculation: 1.60 × 10¹⁴ FLOP/s × 1 GPU × 5.6 × 10⁴ s × 0.4 = 3.6 × 10¹⁸ FLOPs (Rounded to 1.0 × 10¹⁹ FLOPs)
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 GeForce RTX 3090
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
- Closed — provider access only
- Model access
- Hosted access (no API)
- Training code
- Open (non-commercial)
https://github.com/Shihang-Wang-58/DeepSA
How it is classified
Labels the source dataset applies when tracking notable models, and how confident it is in the entry.
- Record confidence
- Speculative
Sources
Where this record came from and when it was last checked.
- Reference
- DeepSA: a deep-learning driven predictor of compound synthesis accessibility
- Last updated
- 28 November 2025
What the numbers mean
Where it came from
DeepSA was published by ShanghaiTech University, in China, in November 2023. It comes out of academia.
It works in Biology, and is recorded as doing synthetic accessibility model.
Its weights were never published, so it can only be reached through its provider. No graphics card changes that.
What went into building it
The training run consumed about 1 × 10¹⁹ FLOP, on NVIDIA GeForce RTX 3090. That figure describes the cost of creating it and has no bearing on how quickly it generates text.
Answers
DeepSA — common questions
When was DeepSA released?
DeepSA was published in November 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.
What is DeepSA used for?
DeepSA works in Biology, and is recorded as handling synthetic accessibility model. These are the areas it was designed around; they describe intent rather than a hard boundary.
How much compute was used to train DeepSA?
Around 1 × 10¹⁹ FLOP, on NVIDIA GeForce RTX 3090. 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.
What GPU do I need to run DeepSA?
None. DeepSA 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.
Is DeepSA open source?
No. DeepSA has not had its weights published, so it exists only as a service controlled by its owner.
How many parameters does DeepSA have?
No parameter count has been published for DeepSA, which is why no memory or speed figure appears on this page.
Who created DeepSA?
DeepSA was published by ShanghaiTech University, based in China, categorised as academia.
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.