fix: Updated README
Browse files
README.md
CHANGED
|
@@ -19,7 +19,9 @@ Join us in <a href="https://discord.gg/3cGQn9b3YM" target="_blank">Discord</a> a
|
|
| 19 |
|
| 20 |
## Disclaimer
|
| 21 |
|
| 22 |
-
This is a f16 GGUF version, original model card is copy & pasted as is below. Followed this [README](https://github.com/OpenBMB/MiniCPM/blob/main/README-en.md#llamacpp) for quantization.
|
|
|
|
|
|
|
| 23 |
|
| 24 |
## Introduction
|
| 25 |
MiniCPM3-4B is the 3rd generation of MiniCPM series. The overall performance of MiniCPM3-4B surpasses Phi-3.5-mini-Instruct and GPT-3.5-Turbo-0125, being comparable with many recent 7B~9B models.
|
|
|
|
| 19 |
|
| 20 |
## Disclaimer
|
| 21 |
|
| 22 |
+
This is a f16 GGUF version, original model card is copy & pasted as is below. Followed this [README](https://github.com/OpenBMB/MiniCPM/blob/main/README-en.md#llamacpp) for quantization and inference.
|
| 23 |
+
|
| 24 |
+
**NOTE:** currently inference only works with the [OpenBMB fork of llama.cpp](https://github.com/OpenBMB/llama.cpp).
|
| 25 |
|
| 26 |
## Introduction
|
| 27 |
MiniCPM3-4B is the 3rd generation of MiniCPM series. The overall performance of MiniCPM3-4B surpasses Phi-3.5-mini-Instruct and GPT-3.5-Turbo-0125, being comparable with many recent 7B~9B models.
|