In short
How they differ.
Worked out from the figures below. Prices and specs are the makers' own, read Oct 20, 2026.
- Both can be downloaded and run on your own machine.
- No independent board we follow ranks both models yet.
Side by side
The specs.
| Gemma 4 26B A4B | Mistral Medium 3.5 | |
|---|---|---|
| Made by | Google · US | Mistral AI · France |
| Released | Apr 2, 2026 | Apr 28, 2026 |
| Price per 1M tokens | Free to download; hosts set prices | $1.50 in / $7.50 out |
| Context window | 256K | 256K |
| Takes in | text, image, video | text, image |
| Download and run it yourself | Yes (Apache 2.0) | Yes (Modified MIT) |
| Size | 25.2B parameters, 3.8B active | 128B (dense) parameters |
| Maker says | “Today, we are introducing Gemma 4 — our most intelligent open models to date.” | “Our frontier-class multimodal model optimized for agentic and coding use cases.” |
| Sources | Prices, Oct 20, 2026 · Description | Prices, Oct 20, 2026 · Description |
Independent tests
Where they place.
None of the boards we follow lists either model yet.
More comparisons