Transformers
mpt
Composer
MosaicML
llm-foundry
StreamingDatasets
text-generation-inference
6 papers
TheBloke commited on
Commit
667ea0e
1 Parent(s): 111dfc1

Update GGML models for latest ggml commit with ggml version included.

Browse files
Files changed (1) hide show
  1. README.md +7 -7
README.md CHANGED
@@ -31,12 +31,12 @@ Please note that these MPT GGMLs are **not compatbile with llama.cpp**. Right no
31
  ## Provided files
32
  | Name | Quant method | Bits | Size | RAM required | Use case |
33
  | ---- | ---- | ---- | ---- | ---- | ----- |
34
- `mpt-7b.ggmlv2.q4_0.bin` | q4_0 | 4bit | 4.16GB | 6.2GB | 4-bit. |
35
- `mpt-7b.ggmlv2.q4_1.bin` | q4_0 | 4bit | 4.99GB | 7.2GB | 4-bit. Higher accuracy than q4_0 but not as high as q5_0. However has quicker inference than q5 models. |
36
- `mpt-7b.ggmlv2.q5_0.bin` | q5_0 | 5bit | 4.57GB | 6.8GB | 5-bit. Higher accuracy, higher resource usage and slower inference. |
37
- `mpt-7b.ggmlv2.q5_1.bin` | q5_1 | 5bit | 4,99GB | 7.2GB | 5-bit. Even higher accuracy, and higher resource usage and slower inference. |
38
- `mpt-7b.ggmlv2.q8_0.bin` | q8_0 | 8bit | 7.48GB | 9.6GB | 8-bit. Almost indistinguishable from float16. Huge resource use and slow. Not recommended for normal use. |
39
- `mpt-7b.ggmlv2.fp16.bin` | fp16 | 16bit | 13.3GB | 15.5GB | Full 16-bit. |
40
 
41
  ## Compatibilty
42
 
@@ -57,7 +57,7 @@ mkdir build
57
  cd build
58
  cmake ..
59
  cmake --build . --config Release
60
- bin/mpt -m /path/to/mpt-7b.ggmlv2.q4_0.bin -t 8 -n 512 -p "Write a story about llamas"
61
  ```
62
 
63
  Please see the ggml repo for other build options.
 
31
  ## Provided files
32
  | Name | Quant method | Bits | Size | RAM required | Use case |
33
  | ---- | ---- | ---- | ---- | ---- | ----- |
34
+ `mpt-7b.ggmlv3.q4_0.bin` | q4_0 | 4bit | 4.16GB | 6.2GB | 4-bit. |
35
+ `mpt-7b.ggmlv3.q4_1.bin` | q4_0 | 4bit | 4.99GB | 7.2GB | 4-bit. Higher accuracy than q4_0 but not as high as q5_0. However has quicker inference than q5 models. |
36
+ `mpt-7b.ggmlv3.q5_0.bin` | q5_0 | 5bit | 4.57GB | 6.8GB | 5-bit. Higher accuracy, higher resource usage and slower inference. |
37
+ `mpt-7b.ggmlv3.q5_1.bin` | q5_1 | 5bit | 4,99GB | 7.2GB | 5-bit. Even higher accuracy, and higher resource usage and slower inference. |
38
+ `mpt-7b.ggmlv3.q8_0.bin` | q8_0 | 8bit | 7.48GB | 9.6GB | 8-bit. Almost indistinguishable from float16. Huge resource use and slow. Not recommended for normal use. |
39
+ `mpt-7b.ggmlv3.fp16.bin` | fp16 | 16bit | 13.3GB | 15.5GB | Full 16-bit. |
40
 
41
  ## Compatibilty
42
 
 
57
  cd build
58
  cmake ..
59
  cmake --build . --config Release
60
+ bin/mpt -m /path/to/mpt-7b.ggmlv3.q4_0.bin -t 8 -n 512 -p "Write a story about llamas"
61
  ```
62
 
63
  Please see the ggml repo for other build options.