DeepSeek R1 Distill 7B
Thinks step by step before answering
Install with Bobbin
Don't have the app? Download it here.
A reasoning model: it works through a problem in stages rather than answering immediately. Noticeably better at maths, logic puzzles and multi-step questions than other models this size.
The trade-off is speed. It writes out its thinking, so answers take longer to arrive. Worth it when the answer matters more than the wait.
Available versions
| Quantization | Size | VRAM | RAM | Context | |
|---|---|---|---|---|---|
| Q4_K_M | 4.7 GB | 5.0 GBest. | 8.8 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.