Mirrors/text-generation-webui
1
0
Fork 0
mirror of https://github.com/oobabooga/text-generation-webui.git synced 2024-11-22 08:07:56 +01:00
Code Issues Packages Projects Releases Wiki Activity

Bump ExLlamaV2 to 0.0.5 (#4186)

Browse Source
This commit is contained in:
turboderp 2023-10-06 00:12:22 +02:00 committed by GitHub
parent 7ffb424c7b
commit 8a98646a21
No known key found for this signature in database
GPG Key ID: 4AEE18F83AFDEB23
11 changed files with 16 additions and 14 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

2
modules/exllamav2.py
View File

@ -110,7 +110,7 @@ class Exllamav2Model:
has_leading_space = False
for i in range(max_new_tokens):
logits = self.model.forward(ids[:, -1:], self.cache, input_mask=None).float().cpu()
token, _ = ExLlamaV2Sampler.sample(logits, settings, ids, random.random())
token, _, _= ExLlamaV2Sampler.sample(logits, settings, ids, random.random(), self.tokenizer)
ids = torch.cat([ids, token], dim=1)
if i == 0 and self.tokenizer.tokenizer.IdToPiece(int(token)).startswith('▁'):

2
modules/loaders.py
View File

@ -216,6 +216,7 @@ loaders_samplers = {
'guidance_scale',
'negative_prompt',
'ban_eos_token',
'add_bos_token',
'custom_token_bans',
'auto_max_new_tokens',
},
@ -228,6 +229,7 @@ loaders_samplers = {
'repetition_penalty_range',
'seed',
'ban_eos_token',
'add_bos_token',
'custom_token_bans',
'auto_max_new_tokens',
},

6
requirements.txt
View File

@ -8,7 +8,7 @@ accelerate==0.23.*
colorama
datasets
einops
exllamav2==0.0.4; platform_system != "Darwin" and platform_machine != "x86_64"
exllamav2==0.0.5; platform_system != "Darwin" and platform_machine != "x86_64"
markdown
numpy==1.24
optimum==1.13.1
@ -40,8 +40,8 @@ https://github.com/PanQiWei/AutoGPTQ/releases/download/v0.4.2/auto_gptq-0.4.2+cu
https://github.com/PanQiWei/AutoGPTQ/releases/download/v0.4.2/auto_gptq-0.4.2+cu117-cp310-cp310-linux_x86_64.whl; platform_system == "Linux" and platform_machine == "x86_64"
https://github.com/jllllll/exllama/releases/download/0.0.17/exllama-0.0.17+cu117-cp310-cp310-win_amd64.whl; platform_system == "Windows"
https://github.com/jllllll/exllama/releases/download/0.0.17/exllama-0.0.17+cu117-cp310-cp310-linux_x86_64.whl; platform_system == "Linux" and platform_machine == "x86_64"
https://github.com/turboderp/exllamav2/releases/download/v0.0.4/exllamav2-0.0.4+cu117-cp310-cp310-win_amd64.whl; platform_system == "Windows"
https://github.com/turboderp/exllamav2/releases/download/v0.0.4/exllamav2-0.0.4+cu117-cp310-cp310-linux_x86_64.whl; platform_system == "Linux" and platform_machine == "x86_64"
https://github.com/turboderp/exllamav2/releases/download/v0.0.5/exllamav2-0.0.5+cu117-cp310-cp310-win_amd64.whl; platform_system == "Windows"
https://github.com/turboderp/exllamav2/releases/download/v0.0.5/exllamav2-0.0.5+cu117-cp310-cp310-linux_x86_64.whl; platform_system == "Linux" and platform_machine == "x86_64"
https://github.com/Dao-AILab/flash-attention/releases/download/v2.3.0/flash_attn-2.3.0+cu117torch2.0cxx11abiFALSE-cp310-cp310-linux_x86_64.whl; platform_system == "Linux" and platform_machine == "x86_64"
https://github.com/jllllll/llama-cpp-python-cuBLAS-wheels/releases/download/textgen-webui/llama_cpp_python_cuda-0.2.11+cu117-cp310-cp310-win_amd64.whl; platform_system == "Windows"
https://github.com/jllllll/llama-cpp-python-cuBLAS-wheels/releases/download/textgen-webui/llama_cpp_python_cuda-0.2.11+cu117-cp310-cp310-manylinux_2_31_x86_64.whl; platform_system == "Linux" and platform_machine == "x86_64"

2
requirements_amd.txt
View File

@ -8,7 +8,7 @@ accelerate==0.23.*
colorama
datasets
einops
exllamav2==0.0.4
exllamav2==0.0.5
markdown
numpy==1.24
optimum==1.13.1

2
requirements_amd_noavx2.txt
View File

@ -8,7 +8,7 @@ accelerate==0.23.*
colorama
datasets
einops
exllamav2==0.0.4
exllamav2==0.0.5
markdown
numpy==1.24
optimum==1.13.1

2
requirements_apple_intel.txt
View File

@ -8,7 +8,7 @@ accelerate==0.23.*
colorama
datasets
einops
exllamav2==0.0.4
exllamav2==0.0.5
markdown
numpy==1.24
optimum==1.13.1

2
requirements_apple_silicon.txt
View File

@ -8,7 +8,7 @@ accelerate==0.23.*
colorama
datasets
einops
exllamav2==0.0.4
exllamav2==0.0.5
markdown
numpy==1.24
optimum==1.13.1

2
requirements_cpu_only.txt
View File

@ -8,7 +8,7 @@ accelerate==0.23.*
colorama
datasets
einops
exllamav2==0.0.4
exllamav2==0.0.5
markdown
numpy==1.24
optimum==1.13.1

2
requirements_cpu_only_noavx2.txt
View File

@ -8,7 +8,7 @@ accelerate==0.23.*
colorama
datasets
einops
exllamav2==0.0.4
exllamav2==0.0.5
markdown
numpy==1.24
optimum==1.13.1

6
requirements_noavx2.txt
View File

@ -8,7 +8,7 @@ accelerate==0.23.*
colorama
datasets
einops
exllamav2==0.0.4; platform_system != "Darwin" and platform_machine != "x86_64"
exllamav2==0.0.5; platform_system != "Darwin" and platform_machine != "x86_64"
markdown
numpy==1.24
optimum==1.13.1
@ -40,8 +40,8 @@ https://github.com/PanQiWei/AutoGPTQ/releases/download/v0.4.2/auto_gptq-0.4.2+cu
https://github.com/PanQiWei/AutoGPTQ/releases/download/v0.4.2/auto_gptq-0.4.2+cu117-cp310-cp310-linux_x86_64.whl; platform_system == "Linux" and platform_machine == "x86_64"
https://github.com/jllllll/exllama/releases/download/0.0.17/exllama-0.0.17+cu117-cp310-cp310-win_amd64.whl; platform_system == "Windows"
https://github.com/jllllll/exllama/releases/download/0.0.17/exllama-0.0.17+cu117-cp310-cp310-linux_x86_64.whl; platform_system == "Linux" and platform_machine == "x86_64"
https://github.com/turboderp/exllamav2/releases/download/v0.0.4/exllamav2-0.0.4+cu117-cp310-cp310-win_amd64.whl; platform_system == "Windows"
https://github.com/turboderp/exllamav2/releases/download/v0.0.4/exllamav2-0.0.4+cu117-cp310-cp310-linux_x86_64.whl; platform_system == "Linux" and platform_machine == "x86_64"
https://github.com/turboderp/exllamav2/releases/download/v0.0.5/exllamav2-0.0.5+cu117-cp310-cp310-win_amd64.whl; platform_system == "Windows"
https://github.com/turboderp/exllamav2/releases/download/v0.0.5/exllamav2-0.0.5+cu117-cp310-cp310-linux_x86_64.whl; platform_system == "Linux" and platform_machine == "x86_64"
https://github.com/Dao-AILab/flash-attention/releases/download/v2.3.0/flash_attn-2.3.0+cu117torch2.0cxx11abiFALSE-cp310-cp310-linux_x86_64.whl; platform_system == "Linux" and platform_machine == "x86_64"
https://github.com/jllllll/llama-cpp-python-cuBLAS-wheels/releases/download/textgen-webui/llama_cpp_python_cuda-0.2.11+cu117-cp310-cp310-win_amd64.whl; platform_system == "Windows"
https://github.com/jllllll/llama-cpp-python-cuBLAS-wheels/releases/download/textgen-webui/llama_cpp_python_cuda-0.2.11+cu117-cp310-cp310-manylinux_2_31_x86_64.whl; platform_system == "Linux" and platform_machine == "x86_64"

2
requirements_nowheels.txt
View File

@ -8,7 +8,7 @@ accelerate==0.23.*
colorama
datasets
einops
exllamav2==0.0.4
exllamav2==0.0.5
markdown
numpy==1.24
optimum==1.13.1