metadata
base_model:
- enkryptai/DeepSeek-R1-Distill-Llama-8B-Enkrypt-Aligned
language:
- en
pipeline_tag: text-generation
Melvin56/DeepSeek-R1-Distill-Llama-8B-Enkrypt-Aligned-GGUF
Original Model : enkryptai/DeepSeek-R1-Distill-Llama-8B-Enkrypt-Aligned
All quants are made using the imatrix option.
| Model | Size (GB) |
|---|---|
| Q2_K | 3.17 |
| Q3_K_M | 4.02 |
| Q4_K_M | 4.92 |
| Q5_K_M | 5.72 |
| Q6_K | 6.59 |
| Q8_0 | 8.54 |
| F16 | 16.2 |
| CPU (AVX2) | CPU (ARM NEON) | Metal | cuBLAS | rocBLAS | SYCL | CLBlast | Vulkan | Kompute | |
|---|---|---|---|---|---|---|---|---|---|
| K-quants | ✅ | ✅ | ✅ | ✅ | ✅ | ✅ | ✅ 🐢5 | ✅ 🐢5 | ❌ |
| I-quants | ✅ 🐢4 | ✅ 🐢4 | ✅ 🐢4 | ✅ | ✅ | Partial¹ | ❌ | ❌ | ❌ |
✅: feature works
🚫: feature does not work
❓: unknown, please contribute if you can test it youself
🐢: feature is slow
¹: IQ3_S and IQ1_S, see #5886
²: Only with -ngl 0
³: Inference is 50% slower
⁴: Slower than K-quants of comparable size
⁵: Slower than cuBLAS/rocBLAS on similar cards
⁶: Only q8_0 and iq4_nl