Add model card metadata
Browse files
README.md
CHANGED
|
@@ -1,3 +1,15 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
# Flint EPQ for llama.cpp
|
| 2 |
|
| 3 |
CPU expert paging for the Unsloth `flint-UD-Q4_K_XL.gguf` checkpoint. Routed gate/up/down expert weights stay on disk and are copied into a bounded RAM arena as needed. Flint's router, selected-expert weights, shared expert, attention and dense layers remain on the upstream llama.cpp graph.
|
|
|
|
| 1 |
+
---
|
| 2 |
+
library_name: llama.cpp
|
| 3 |
+
license: mit
|
| 4 |
+
base_model: RecursiveSI/flint
|
| 5 |
+
tags:
|
| 6 |
+
- flint
|
| 7 |
+
- llama.cpp
|
| 8 |
+
- moe
|
| 9 |
+
- expert-offloading
|
| 10 |
+
- cpu-inference
|
| 11 |
+
---
|
| 12 |
+
|
| 13 |
# Flint EPQ for llama.cpp
|
| 14 |
|
| 15 |
CPU expert paging for the Unsloth `flint-UD-Q4_K_XL.gguf` checkpoint. Routed gate/up/down expert weights stay on disk and are copied into a bounded RAM arena as needed. Flint's router, selected-expert weights, shared expert, attention and dense layers remain on the upstream llama.cpp graph.
|