unsloth
DeepSeek R1 Distill Qwen 1.5B
DeepSeek R1 Distill Qwen 1.5B is a 1.78B dense qwen2 model from unsloth. Configured context length is 131,072 tokens. License: apache-2.0. Quant sizes are the GGUF file bytes published on Hugging Face when a file exists.
Parameters1.8B
Context128K
Attentiongqa
Licenseapache-2.0
28 layers · 2 KV heads · head dim 128 · GGUF repo · config
Quant file sizes
| Quant | File size | Source |
|---|---|---|
| BF16 | 3.32 GB | Hugging Face file bytes |
| Q8_0 | 1.76 GB | Hugging Face file bytes |
| Q6_K | 1.36 GB | Hugging Face file bytes |
| Q5_K_M | 1.20 GB | Hugging Face file bytes |
| UD-Q4_K_XL | 1.11 GB | Hugging Face file bytes |
| Q4_K_M | 1.04 GB | Hugging Face file bytes |
| UD-IQ4_XS | 1.02 GB | Hugging Face file bytes |
| UD-Q3_K_XL | 0.90 GB | Hugging Face file bytes |
| Q3_K_M | 0.86 GB | Hugging Face file bytes |
| UD-IQ3_XXS | 0.82 GB | Hugging Face file bytes |
| UD-Q2_K_XL | 0.82 GB | Hugging Face file bytes |
| Q2_K | 1.45 GB | Hugging Face file bytes |
| UD-IQ2_XXS | 0.72 GB | Hugging Face file bytes |
| UD-IQ1_M | 0.67 GB | Hugging Face file bytes |
| UD-IQ1_S | 0.65 GB | Hugging Face file bytes |
On common machines
| Device | Verdict | Quant | Speed |
|---|---|---|---|
| GeForce RTX 4060 Ti 8GB | Runs great | BF16 | 42–57 |
| GeForce RTX 3060 12GB | Runs great | BF16 | 52–71 |
| GeForce RTX 4060 Ti 16GB | Runs great | BF16 | 42–57 |
| GeForce RTX 4090 | Runs great | BF16 | 146–199 |
| GeForce RTX 5090 | Runs great | BF16 | 260–354 |
| Apple M4 Pro | Runs great | BF16 | 40–54 |
| Apple M5 | Runs great | BF16 | 22–30 |
| CPU only · DDR5-5600 | Runs on CPU | BF16 | 13–18 |
Smallest devices that fit it fully
- GeForce RTX 3050 6GB — Runs, BF16, 24–33 tok/s
- GeForce RTX 2060 6GB — Runs, BF16, 49–66 tok/s
- GeForce GTX 1660 Super — Runs, BF16, 49–66 tok/s
- GeForce GTX 1660 Ti — Runs, BF16, 42–57 tok/s
- Intel Arc A380 — Runs, BF16, 27–37 tok/s
- GeForce RTX 5060 Ti 8GB — Runs great, BF16, 65–89 tok/s
Run it
Commands for the GeForce RTX 4060 Ti 16GB estimate (BF16).
llama.cpp
llama-server -hf unsloth/DeepSeek-R1-Distill-Qwen-1.5B-GGUF:BF16 -c 8192 -ngl 28
No matching name in the Ollama library snapshot.
LM Studio
lms get unsloth/DeepSeek-R1-Distill-Qwen-1.5B-GGUF # Then load quant BF16