Update README.md
Browse files
README.md
CHANGED
|
@@ -21,7 +21,7 @@ tags:
|
|
| 21 |
# Quant Infos
|
| 22 |
|
| 23 |
- Not supported in llama.cpp master; Requires the latest version of the phi3 128k [branch](https://github.com/ggerganov/llama.cpp/pull/7225)
|
| 24 |
-
- quants & imatrix are still in the oven will follow soon TM
|
| 25 |
<!-- - quants done with an importance matrix for improved quantization loss -->
|
| 26 |
<!-- - gguf & imatrix generated from bf16 for "optimal" accuracy loss (some say this is snake oil, but it can't hurt) -->
|
| 27 |
<!-- - Wide coverage of different gguf quant types from Q\_8\_0 down to IQ1\_S -->
|
|
|
|
| 21 |
# Quant Infos
|
| 22 |
|
| 23 |
- Not supported in llama.cpp master; Requires the latest version of the phi3 128k [branch](https://github.com/ggerganov/llama.cpp/pull/7225)
|
| 24 |
+
- just bf16 for now, quants & imatrix are still in the oven will follow soon TM
|
| 25 |
<!-- - quants done with an importance matrix for improved quantization loss -->
|
| 26 |
<!-- - gguf & imatrix generated from bf16 for "optimal" accuracy loss (some say this is snake oil, but it can't hurt) -->
|
| 27 |
<!-- - Wide coverage of different gguf quant types from Q\_8\_0 down to IQ1\_S -->
|