Replit Code Models by Replit
About
The Replit Code family comprises a set of large language models (LLMs) adept in code completion, aiding developers by generating and completing code snippets, and providing code suggestions. Trained on extensive datasets, including open-source code and StackExchange data, these models support multiple programming languages. Replit emphasizes open-source principles, making these models accessible on platforms like Hugging Face to foster community engagement and further innovation. The Code family offers models of varying scales, with newer versions demonstrating enhanced performance and expanded language support. However, like other LLMs, they may occasionally produce inappropriate content based on their training data 2311.
Current Variants
Use-when guidance is derived from seed capabilities, context, release, and replacement fields.
| Model | Use when | Released | Signals | Status |
|---|---|---|---|---|
| Replit Code | Use when the workload needs 3B parameters. | 2023-06 | 3B parameters | Current |
| Replit Code 1.5 | Use when the workload needs 3B parameters. | 2023-06 | 3B parameters | Current |
Release Timeline
1 release groupSpecifications(2 models)
| Model | Released | Parameters |
|---|---|---|
| Replit Code | 2023-06 | 3B |
| Replit Code 1.5 | 2023-06 | 3B |
Available From(1 provider)
Frequently Asked Questions
- What is Replit Code used for?
- Replit Code is used for coding. The family description and listed model capabilities point to those workloads as the best fit.
- How does Replit Code compare to StarCoder?
- Replit Code by Replit is strongest where you need coding, while StarCoder by ServiceNow Research is the closest related family to check for coding. Replit Code has 2 listed variants, while StarCoder reaches up to 8K context, so compare the specs and pricing tables before choosing a production model.
- Which Replit Code model should I use?
- If price is the main constraint, use the pricing table first because Replit Code does not have complete provider pricing in the local data. For the most capable/latest local choice, evaluate Replit Code.
