DeepSeek R1 (8B, Q4)
Reasoning-tuned 8B model. Slower than a plain chat model, but thinks through its answers before replying.
| At a glance | |
|---|---|
| Made by | DeepSeek |
| Download size | 5.2 GB |
| Parameters | 8B |
| Quantization | Q4_K_M |
| Context window | 128K tokens |
| Uses tools | Yes |
| Best for | Reasoning |
| Runs on | llama.cpp, MLX, Ollama |
| License | MIT |
| Added to the catalog | 2026-08-15 |
Gezel downloads this model the first time a gezel needs it, checks the file against its published hash, and applies tuned settings for it.