AI models
PolyCoder
Training compute
1.1×10²¹ FLOP
Parameters
2.7B
Published
Feb 26, 2022
PolyCoder is an AI model developed by Carnegie Mellon University (CMU) (United States), first published in February 2022. It works in the language domain, on tasks such as code generation.
Training it took an estimated 1.1×10²¹ FLOP of compute (estimation method: hardware). The model has 2,700,000,000 parameters. It was trained on roughly 39.3B datapoints. Training ran on NVIDIA Quadro RTX 8000 for about 1K hours.
Access: Open weights (unrestricted). Its weights are openly available. The reference paper has 860 citations. Epoch AI rates the confidence of this record as likely.
Full record
- Organization
- Carnegie Mellon University (CMU)
- Country of organization
- United States
- Domain
- Language
- Task
- Code generation
- Training compute
- 1.1×10²¹ FLOP
- Compute estimation method
- Hardware
- Parameters
- 2,700,000,000
- Dataset size
- 39.3B
- Training hardware
- NVIDIA Quadro RTX 8000
- Training time
- 1,000 h
- Numerical format
- FP16
- Model accessibility
- Open weights (unrestricted)
- Open weights
- Yes
- Citations
- 860
- Epoch confidence
- Likely
More from Carnegie Mellon University (CMU)
SourceEpoch AI, 'AI Models'. Published online at epoch.ai. Retrieved 2026-07-29 from https://epoch.ai/data/ai-models. Licensed under CC BY 4.0.
← All ai models