root/llama.cpp
1
0
Fork 0
mirror of https://github.com/ggerganov/llama.cpp.git synced 2024-11-14 06:49:54 +00:00
Code Issues Actions 14 Packages Projects Releases Wiki Activity

Updated Feature matrix (markdown)

Romain D 2024-03-05 12:38:35 +00:00
parent 8abf888c87
commit a659cd1217
1 changed files with 5 additions and 3 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

6
Feature-matrix.md

@ -3,7 +3,9 @@
| **K-quants** | ✅ | ✅ | ✅ | ✅ | ✅ | ✅ | ✅ | ✅ | 🚫 |
| **I-quants** | ✅ (SLOW) | ✅ (SLOW) | ✅ (SLOW) | ✅ | ✅ | Partial¹ | 🚫 | 🚫 | 🚫 |
| **Multi-GPU** | N/A | N/A | N/A | ✅ | ❓ | 🚫 | ❓ | ✅ | ❓ |
| **K cache quants** | ✅ | ❓ | ❓ | ✅ | Only q8_0 (SLOW) | ❓ | ✅ | 🚫 | 🚫 |
| **MoE architecture** | ✅ | ❓ | ✅ | ✅ | ✅ | ❓ | Only -ngl 0 | 🚫 | 🚫 |
| **K cache quants** | ✅ | ❓ | ❓ | ✅ | Partial³ (SLOW) | ❓ | ✅ | 🚫 | 🚫 |
| **MoE architecture** | ✅ | ❓ | ✅ | ✅ | ✅ | ❓ | Partial² | 🚫 | 🚫 |
* ¹: IQ3_S and IQ1_S, see #5886
* ²: Only with `-ngl 0`
* ³: Only `-ctk q8_0`