New GGMLv3 format for breaking llama.cpp change May 19th commit 2d5db48

README.md

@@ -17,21 +17,26 @@ This repo contains 4bit and 5bit quantised GGML files for CPU inference using [l
 * [4bit and 5bit GGML models for CPU inference in llama.cpp](https://huggingface.co/TheBloke/gpt4-alpaca-lora_mlp-65B-GGML)
 * [float16 unquantised model for GPU inference and further conversions](https://huggingface.co/TheBloke/gpt4-alpaca-lora_mlp-65B-HF)
 
-## REQUIRES LATEST LLAMA.CPP (May 12th 2023 - commit b9fd7ee)!
+## THE FILES IN MAIN BRANCH REQUIRES LATEST LLAMA.CPP (May 19th 2023 - commit 2d5db48)!
 
-llama.cpp recently made a breaking change to its quantisation methods.
+llama.cpp recently made another breaking change to its quantisation methods - https://github.com/ggerganov/llama.cpp/pull/1508
 
-These GGML files were quantised with this the latest llama.cpp code. Therefore you will require llama.cpp compiled on May 12th or later (commit `b9fd7ee` or later) to use them.
+I have quantised the GGML files in this repo with the latest version. Therefore you will require llama.cpp compiled on May 19th or later (commit `2d5db48` or later) to use them.
+
+For files compatible with the previous version of llama.cpp, please see branch `previous_llama_ggmlv2`.
 
 ## Provided files
 | Name | Quant method | Bits | Size | RAM required | Use case |
 | ---- | ---- | ---- | ---- | ---- | ----- |
-`gpt4-alpaca-lora_mlp-65B.ggml.q4_0.bin` | q4_0 | 4bit | 40.8GB | 43GB | 4-bit. |
-`gpt4-alpaca-lora_mlp-65B.ggml.q5_0.bin` | q5_0 | 5bit | 44.9GB | 47GB | 5-bit. Higher accuracy, higher resource usage and slower inference.  |
-`gpt4-alpaca-lora_mlp-65B.ggml.q5_1.bin` | q5_1 | 5bit | 49.0GB | 51GB | 5-bit. Even higher accuracy, higher resource usage and slower inference. |
+`gpt4-alpaca-lora_mlp-65B.ggmlv3.q4_0.bin` | q4_0 | 4bit | 40.8GB | 43GB | 4-bit. |
+`gpt4-alpaca-lora_mlp-65B.ggmlv3.q4_1.bin` | q4_0 | 4bit | 44.9GB | 47GB | 4-bit. Higher accuracy than q4_0 but not as high as q5_0. However has quicker inference than q5 models. |
+`gpt4-alpaca-lora_mlp-65B.ggmlv3.q5_0.bin` | q5_0 | 5bit | 44.9GB | 47GB | 5-bit. Higher accuracy, higher resource usage and slower inference.  |
+`gpt4-alpaca-lora_mlp-65B.ggmlv3.q5_1.bin` | q5_1 | 5bit | 49.0GB | 51GB | 5-bit. Even higher accuracy, higher resource usage and slower inference. |
 
 Note: no q8_0 will be provided as HF won't allow uploading of files larger than 50GB :)
 
+I am investigating other methods, eg a split ZIP file, and will try to upload this soon.
+
 # Original model card
 
 This repo provides the training checkpoint of LLaMA on the alpaca_data_gpt4 dataset via LoRA [MLP] on 8xA100(80G). 
@@ -155,4 +160,4 @@ for i in range(2,10):
 
 ---
 
-> [1] Junxian He, Chunting Zhou, Xuezhe Ma, Taylor Berg-Kirkpatrick, Graham Neubig: Towards a Unified View of Parameter-Efficient Transfer Learning. ICLR 2022
+> [1] Junxian He, Chunting Zhou, Xuezhe Ma, Taylor Berg-Kirkpatrick, Graham Neubig: Towards a Unified View of Parameter-Efficient Transfer Learning. ICLR 2022