Phi-2 Models by Microsoft Research
Details
Capabilities
About
The Phi family of language models, developed by Microsoft Research, comprises several small language models (SLMs) designed to achieve high performance despite their relatively small size. These models employ a Transformer architecture and are trained on a blend of synthetic and web datasets 12. Emphasizing the quality of training data, the models prioritize "textbook-quality" information to enhance reasoning and understanding capabilities 1. The Phi series includes Phi-1, Phi-1.5, and Phi-2, with each version incorporating advancements in model scaling and data curation 1. Phi-2, the latest in the series, has 2.7 billion parameters and exhibits state-of-the-art performance among base models with fewer than 13 billion parameters across various benchmarks 12. Notably, Phi-2 has not been subjected to reinforcement learning from human feedback (RLHF) or instruction fine-tuning 12, and the models are accessible to researchers to aid exploration of safety challenges and other language model developments 2.
Current Variants
Use-when guidance is derived from seed capabilities, context, release, and replacement fields.
Use when the workload needs 2k context, 2.7B parameters, and structured outputs.
| Model | Use when | Released | Signals | Status |
|---|---|---|---|---|
| Phi-2 | Use when the workload needs 2k context, 2.7B parameters, and structured outputs. | 2023-12 | 2k context2.7B parametersstructured outputs | Current |
Release Timeline
1 release groupSpecifications(1 models)
| Model | Released | Context | Parameters | Structured Outputs |
|---|---|---|---|---|
| Phi-2 | 2023-12 | 2k | 2.7B | Yes |
Available From(5 providers)
Pricing
| Model | Provider | Input / 1M | Output / 1M | Type |
|---|---|---|---|---|
| Phi-2 | Replicate API | $0.05 | $0.25 | Serverless |
| Phi-2 | Microsoft Foundry | $0.07 | $0.07 | Provisioned |
| Phi-2 | Together AI | $0.1 | $0.1 | Serverless |
| Phi-2 | Fireworks AI | $0.1 | $0.1 | Provisioned |
Frequently Asked Questions
- What is Phi-2 used for?
- Phi-2 is used for structured outputs. The family description and listed model capabilities point to those workloads as the best fit.
- How does Phi-2 compare to Harrier?
- Phi-2 by Microsoft Research is strongest where you need structured outputs, while Harrier by Microsoft Research is the closest related family to check for embedding. Phi-2 has 1 listed variant and reaches up to 2k context, while Harrier reaches up to 33k context, so compare the specs and pricing tables before choosing a production model.
