Phi-4 14B
Trained for reasoning rather than trivia
Install with Bobbin
Don't have the app? Download it here.
Microsoft's 14B model, trained largely on synthetic reasoning data. Stronger on maths and structured problems than general models its size, and weaker on obscure facts.
A good fit for analysis rather than recall.
Available versions
| Quantization | Size | VRAM | RAM | Context | |
|---|---|---|---|---|---|
| Q4_K_M | 9.1 GB | 9.1 GBest. | 14.6 GB | 8k | Install |
Memory needed grows with the length of a conversation, so these figures are for the context shown. Anything marked est. is worked out from the file size rather than measured on real hardware — the app says so there too. The desktop app checks all of it against your actual machine before installing.