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

README.md

@@ -14,7 +14,6 @@ language:
 - en
 library_name: transformers
 pipeline_tag: text-generation
-inference: false
 ---
 
 # Manticore 13B GGML
@@ -25,24 +24,26 @@ This repo is the result of quantising to 4-bit, 5-bit and 8-bit GGML for CPU (+C
 
 ## Repositories available
 
-* [4-bit GPTQ models for GPU inference](https://huggingface.co/TheBloke/Manticore-13B-GPTQ).
-* [4-bit, 5-bit 8-bit GGML models for llama.cpp CPU (+CUDA) inference](https://huggingface.co/TheBloke/Manticore-13B-GGML).
+* [4-bit GPTQ models for GPU inference](https://huggingface.co/TheBloke/TheBloke/Manticore-13B-GPTQ).
+* [4-bit, 5-bit 8-bit GGML models for llama.cpp CPU (+CUDA) inference](https://huggingface.co/TheBloke/TheBloke/Manticore-13B-GGML).
 * [OpenAccess AI Collective's original float16 HF format repo for GPU inference and further conversions](https://huggingface.co/openaccess-ai-collective/manticore-13b).
 
-## THE FILES IN MAIN BRANCH 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
 
 I have quantised the GGML files in this repo with the latest version. Therefore you will require llama.cpp compiled on May 12th or later (commit `b9fd7ee` 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 |
 | ---- | ---- | ---- | ---- | ---- | ----- |
-`manticore-13B.ggmlv2.q4_0.bin` | q4_0 | 4bit | 8.14GB | 10.5GB | 4-bit. |
-`manticore-13B.ggmlv2.q4_1.bin` | q4_1 | 4bit | 8.14GB | 10.5GB | 4-bit. Higher accuracy than q4_0 but not as high as q5_0. However has quicker inference than q5 models. |
-`manticore-13B.ggmlv2.q5_0.bin` | q5_0 | 5bit | 8.95GB | 11.0GB | 5-bit. Higher accuracy, higher resource usage and slower inference. |
-`manticore-13B.ggmlv2.q5_1.bin` | q5_1 | 5bit | 9.76GB | 12.25GB | 5-bit. Even higher accuracy, and higher resource usage and slower inference. |
-`manticore-13B.ggmlv2.q8_0.bin` | q8_0 | 8bit | 14.6GB | 17GB | 8-bit. Almost indistinguishable from float16. Huge resource use and slow. Not recommended for normal use. |
+`manticore-13B.ggmlv3.q4_0.bin` | q4_0 | 4bit | 8.14GB | 10.5GB | 4-bit. |
+`manticore-13B.ggmlv3.q4_1.bin` | q4_0 | 4bit | 8.14GB | 10.5GB | 4-bit. Higher accuracy than q4_0 but not as high as q5_0. However has quicker inference than q5 models. |
+`manticore-13B.ggmlv3.q5_0.bin` | q5_0 | 5bit | 8.95GB | 11.0GB | 5-bit. Higher accuracy, higher resource usage and slower inference. |
+`manticore-13B.ggmlv3.q5_1.bin` | q5_1 | 5bit | 9.76GB | 12.25GB | 5-bit. Even higher accuracy, and higher resource usage and slower inference. |
+`manticore-13B.ggmlv3.q8_0.bin` | q8_0 | 8bit | 14.6GB | 17GB | 8-bit. Almost indistinguishable from float16. Huge resource use and slow. Not recommended for normal use. |
 
 ## How to run in `llama.cpp`