Gemma or Phi: small models for modest hardware
Both lines are built for a single machine, both are in Ollama and neither needs a GPU. Gemma goes down to 270M and covers mobile devices, with CodeGemma for code and FunctionGemma 270M for function calling, but its licensing splits by version: Gemma 4 and DiffusionGemma are Apache 2.0, while earlier releases, CodeGemma and FunctionGemma fall under the Gemma Terms of Use. Phi is MIT throughout and the catalog lists it as supporting Russian, which Gemma does not claim. Phi also has vision and speech versions and reaches 42B-A6.6B, while Gemma was updated more recently, June 2026 against January 2026.
Comparison based on catalog data
| Parameter | Gemma | Phi |
|---|---|---|
| Category | Text, Image + text, Code | Text, Image + text |
| Developer | Google, USA | Microsoft, USA |
| Releases | Feb 2024 – Jun 2026 | Sep 2023 – Jan 2026 |
| Sizes | 270M – 31B | 1.3B – 42B-A6.6B |
| Hardware | Laptop, 1 GPU | Laptop, 1 GPU |
| Commercial use | Commercial use allowed | Commercial use allowed |
| License | Gemma 4 and DiffusionGemma: Apache 2.0; earlier versions, CodeGemma and FunctionGemma: Gemma Terms of Use | MIT |
| Russian | Not stated | Supported |
| Ollama | Yes | Yes |
| Without GPU | Yes | Yes |
| Tasks |
|
|
Choose Gemma if
- You need something truly tiny: Gemma starts at 270M
- The model goes on a phone, a mini PC or other low-end hardware
- You need to read photos of documents and receipts
Choose Phi if
- You want MIT across every version with no terms to parse
- Your assistant must work in Russian, which the catalog lists for Phi
- Your tasks involve reasoning and calculation, education or analytics
Other comparisons
- Gemma or Llama: which to choose for business
- MiniCPM or Gemma: which model to put on the device
- Qwen or GigaChat: which to choose for business
- Qwen or Llama: which to choose for business
- GigaChat or YandexGPT: which to choose for business
- DeepSeek or Qwen: which to choose for business
- Mistral or Llama: which to choose for business
- gpt-oss or Qwen: which one to run on your own server
- Llama or DeepSeek: which to run inside your own perimeter
- Mistral or Qwen: which to choose for business
- YandexGPT or Qwen: which to choose for Russian
- T-Pro or GigaChat: which one for a Russian company
Need a model for your task?
An open model can run on your own server: data stays in-house, there is no per-request fee, and the model can be fine-tuned on your documents.
- SelectThe model and size for your task and hardware budget
- DeployOn your server or in a closed network, with an API
- Fine-tuneOn your data, or connect a knowledge base
- IntegrateInto your CRM, ERP, bot, website or team chat