Mirrors/text-generation-webui
1
0
Fork 0
mirror of https://github.com/oobabooga/text-generation-webui.git synced 2024-11-21 23:57:58 +01:00
Code Issues Packages Projects Releases Wiki Activity

Experimental jank multiGPU inference that's 2x faster than native somehow (#2100)

Browse Source
This commit is contained in:
Alex "mcmonkey" Goodwin 2023-05-17 06:41:09 -07:00 committed by GitHub
parent fd743a0207
commit 1f50dbe352
No known key found for this signature in database
GPG Key ID: 4AEE18F83AFDEB23
4 changed files with 10 additions and 3 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

2
README.md
View File

@ -240,7 +240,7 @@ Optionally, you can use the following command-line flags:
| `--wbits WBITS` | Load a pre-quantized model with specified precision in bits. 2, 3, 4 and 8 are supported. |
| `--model_type MODEL_TYPE` | Model type of pre-quantized model. Currently LLaMA, OPT, and GPT-J are supported. |
| `--groupsize GROUPSIZE` | Group size. |
| `--pre_layer PRE_LAYER` | The number of layers to allocate to the GPU. Setting this parameter enables CPU offloading for 4-bit models. |
| `--pre_layer PRE_LAYER [PRE_LAYER ...]` | The number of layers to allocate to the GPU. Setting this parameter enables CPU offloading for 4-bit models. For multi-gpu, write the numbers separated by spaces, eg `--pre_layer 30 60`. |
| `--checkpoint CHECKPOINT` | The path to the quantized checkpoint file. If not specified, it will be automatically detected. |
| `--monkey-patch` | Apply the monkey patch for using LoRAs with quantized models.
| `--quant_attn` | (triton) Enable quant attention. |

2
docs/GPTQ-models-(4-bit-mode).md
View File

@ -107,6 +107,8 @@ This is the performance:
Output generated in 123.79 seconds (1.61 tokens/s, 199 tokens)
```
You can also use multiple GPUs with `pre_layer` if using the oobabooga fork of GPTQ, eg `--pre_layer 30 60` will load a LLaMA-30B model half onto your first GPU and half onto your second, or `--pre_layer 20 40` will load 20 layers onto GPU-0, 20 layers onto GPU-1, and 20 layers offloaded to CPU.
## Using LoRAs in 4-bit mode
At the moment, this feature is not officially supported by the relevant libraries, but a patch exists and is supported by this web UI: https://github.com/johnsmith0031/alpaca_lora_4bit

7
modules/GPTQ_loader.py
View File

@ -172,7 +172,12 @@ def load_quantized(model_name):
# qwopqwop200's offload
if model_type == 'llama' and shared.args.pre_layer:
model = load_quant(str(path_to_model), str(pt_path), shared.args.wbits, shared.args.groupsize, shared.args.pre_layer)
if len(shared.args.pre_layer) == 1:
pre_layer = shared.args.pre_layer[0]
else:
pre_layer = shared.args.pre_layer
model = load_quant(str(path_to_model), str(pt_path), shared.args.wbits, shared.args.groupsize, pre_layer)
else:
threshold = False if model_type == 'gptj' else 128
model = load_quant(str(path_to_model), str(pt_path), shared.args.wbits, shared.args.groupsize, kernel_switch_threshold=threshold)

2
modules/shared.py
View File

@ -130,7 +130,7 @@ parser.add_argument('--n-gpu-layers', type=int, default=0, help='Number of layer
parser.add_argument('--wbits', type=int, default=0, help='Load a pre-quantized model with specified precision in bits. 2, 3, 4 and 8 are supported.')
parser.add_argument('--model_type', type=str, help='Model type of pre-quantized model. Currently LLaMA, OPT, and GPT-J are supported.')
parser.add_argument('--groupsize', type=int, default=-1, help='Group size.')
parser.add_argument('--pre_layer', type=int, default=0, help='The number of layers to allocate to the GPU. Setting this parameter enables CPU offloading for 4-bit models.')
parser.add_argument('--pre_layer', type=int, nargs="+", help='The number of layers to allocate to the GPU. Setting this parameter enables CPU offloading for 4-bit models. For multi-gpu, write the numbers separated by spaces, eg --pre_layer 30 60.')
parser.add_argument('--checkpoint', type=str, help='The path to the quantized checkpoint file. If not specified, it will be automatically detected.')
parser.add_argument('--monkey-patch', action='store_true', help='Apply the monkey patch for using LoRAs with quantized models.')
parser.add_argument('--quant_attn', action='store_true', help='(triton) Enable quant attention.')