Alibaba's open-weight model family, strong on multilingual and coding tasks.
| Feature | Qwen | Phi |
|---|---|---|
| Context window | 128k tokens | 128k tokens |
| License | Apache-2.0 for most sizes | MIT |
| Open weights | Yes | Yes |
| Parameters | 0.5B to 72B | 3.8B, 14B |
| Provider | Alibaba | Microsoft |
Shaded rows are identical for both; the differing rows are the ones that drive the recommendation below.
Generated from the differences above by structured rules, not by a model.
model · Models
Qwen is an open-weight language model family from Alibaba, released in a range of sizes and with a strong showing on multilingual and code benchmarks. It solves the same problem as other open-weight models: letting a team run a capable model without depending on a closed API. It exists as one of the few families that publishes both very small and very large variants, which makes it a common default when you need a model that fits a specific memory budget.
model · Models
Phi is a family of small language models trained on a deliberately narrow, high-quality corpus of textbooks and exercises rather than on a web-scale crawl. It solves the assumption that capability requires scale, showing that data selection can substitute for a large part of it. It exists because it reframes the question from how much data to what kind of data, and the answer is useful when the budget is a single GPU.
