Dolly 2.0 Models by Databricks Mosaic
About
Dolly 2.0 is a series of open-source large language models (LLMs) created by Databricks, based on the EleutherAI Pythia model family. These models are designed to follow instructions, enabling them to understand and complete natural language tasks effectively. Uniquely licensed for commercial use, Dolly 2.0 provides enterprises the opportunity to leverage these models without restrictive licenses. Available in three different sizes—3 billion, 7 billion, and 12 billion parameters—they offer scalability according to computational needs and performance goals. The training involves a fine-tuning process on the databricks-dolly-15k dataset, a carefully crafted instruction set by Databricks employees. While they may not represent the cutting edge of technology, Dolly 2.0 models display commendable instruction-following abilities, making them a practical choice for various applications.
Current Variants
Use-when guidance is derived from seed capabilities, context, release, and replacement fields.
| Model | Use when | Released | Signals | Status |
|---|---|---|---|---|
| Dolly 2.0 12B | Use when the workload needs 12B parameters. | 2023-04 | 12B parameters | Current |
| Dolly 2.0 7B | Use when the workload needs 7B parameters. | 2023-04 | 7B parameters | Current |
| Dolly 2.0 3B | Use when the workload needs 3B parameters. | 2023-04 | 3B parameters | Current |
Release Timeline
1 release groupSpecifications(3 models)
| Model | Released | Parameters |
|---|---|---|
| Dolly 2.0 12B | 2023-04 | 12b |
| Dolly 2.0 7B | 2023-04 | 7b |
| Dolly 2.0 3B | 2023-04 | 3b |
Available From(2 providers)
Pricing
| Model | Provider | Input / 1M | Output / 1M | Type |
|---|---|---|---|---|
| Dolly 2.0 12B | Microsoft Foundry | $0.07 | $0.07 | Provisioned |
| Dolly 2.0 12B | Replicate API | $0.1 | $0.5 | Serverless |
Frequently Asked Questions
- What is Dolly 2.0 used for?
- Dolly 2.0 is used for math-heavy prompts and chatbot and role-playing use cases. The family description and listed model capabilities point to those workloads as the best fit.
- How does Dolly 2.0 compare to MPT?
- Dolly 2.0 by Databricks Mosaic is strongest where you need math-heavy prompts, while MPT by Databricks Mosaic is the closest related family to check for coding. Dolly 2.0 has 3 listed variants, while MPT reaches up to 8k context, so compare the specs and pricing tables before choosing a production model.
- Which Dolly 2.0 model should I use?
- For the lowest listed input price, start with Dolly 2.0 12B through Microsoft Foundry at $0.07/1M input tokens. For the most capable/latest local choice, evaluate Dolly 2.0 12B.
