Merge pull request #5379 from oobabooga/dev

Merge dev branch

README.md

@@ -415,7 +415,3 @@ If you would like to contribute to the project, check out the [Contributing guid
 ## Acknowledgment
 
 In August 2023, [Andreessen Horowitz](https://a16z.com/) (a16z) provided a generous grant to encourage and support my independent work on this project. I am **extremely** grateful for their trust and recognition.
-
-## Support
-
-[![ko-fi](https://ko-fi.com/img/githubbutton_sm.svg)](https://ko-fi.com/B0B5JFPBO)

docs/12 - OpenAI API.md

@@ -99,7 +99,7 @@ curl http://127.0.0.1:5000/v1/chat/completions \
 
 #### Logits
 
-```
+```shell
 curl -k http://127.0.0.1:5000/v1/internal/logits \
   -H "Content-Type: application/json" \
   -d '{
@@ -110,7 +110,7 @@ curl -k http://127.0.0.1:5000/v1/internal/logits \
 
 #### Logits after sampling parameters
 
-```
+```shell
 curl -k http://127.0.0.1:5000/v1/internal/logits \
   -H "Content-Type: application/json" \
   -d '{

modules/exllamav2.py

@@ -93,17 +93,27 @@ class Exllamav2Model:
 
     def generate_with_streaming(self, prompt, state):
         settings = ExLlamaV2Sampler.Settings()
+
+        settings.token_repetition_penalty = state['repetition_penalty']
+        settings.token_repetition_range = -1 if state['repetition_penalty_range'] <= 0 else state['repetition_penalty_range']
+
+        settings.token_frequency_penalty = state['frequency_penalty']
+        settings.token_presence_penalty = state['presence_penalty']
+
         settings.temperature = state['temperature']
         settings.top_k = state['top_k']
         settings.top_p = state['top_p']
+        settings.top_a = state['top_a']
         settings.min_p = state['min_p']
         settings.tfs = state['tfs']
         settings.typical = state['typical_p']
+
+        settings.temperature_last = state['temperature_last']
+
         settings.mirostat = state['mirostat_mode'] == 2
         settings.mirostat_tau = state['mirostat_tau']
         settings.mirostat_eta = state['mirostat_eta']
-        settings.token_repetition_penalty = state['repetition_penalty']
+
-        settings.token_repetition_range = -1 if state['repetition_penalty_range'] <= 0 else state['repetition_penalty_range']
         if state['ban_eos_token']:
             settings.disallow_tokens(self.tokenizer, [self.tokenizer.eos_token_id])
 

modules/loaders.py

@@ -205,12 +205,16 @@ loaders_samplers = {
     'HQQ': transformers_samplers(),
     'ExLlamav2': {
         'temperature',
+        'temperature_last',
         'top_p',
         'min_p',
         'top_k',
         'typical_p',
         'tfs',
+        'top_a',
         'repetition_penalty',
+        'presence_penalty',
+        'frequency_penalty',
         'repetition_penalty_range',
         'seed',
         'mirostat_mode',

modules/models.py

@@ -162,7 +162,7 @@ def huggingface_loader(model_name):
 
     # DeepSpeed ZeRO-3
     elif shared.args.deepspeed:
-        model = LoaderClass.from_pretrained(path_to_model, torch_dtype=params['torch_dtype'])
+        model = LoaderClass.from_pretrained(path_to_model, torch_dtype=params['torch_dtype'], trust_remote_code=params['trust_remote_code'])
         model = deepspeed.initialize(model=model, config_params=ds_config, model_parameters=None, optimizer=None, lr_scheduler=None)[0]
         model.module.eval()  # Inference
         logger.info(f'DeepSpeed ZeRO-3 is enabled: {is_deepspeed_zero3_enabled()}')

requirements.txt

@@ -2,9 +2,9 @@ accelerate==0.25.*
 colorama
 datasets
 einops
-exllamav2==0.0.11; platform_system != "Darwin" and platform_machine != "x86_64"
+exllamav2==0.0.12; platform_system != "Darwin" and platform_machine != "x86_64"
 gradio==3.50.*
-hqq==0.1.2
+hqq==0.1.2.post1
 jinja2==3.1.2
 lm_eval==0.3.0
 markdown
@@ -67,14 +67,14 @@ https://github.com/jllllll/AutoGPTQ/releases/download/v0.6.0/auto_gptq-0.6.0+cu1
 https://github.com/jllllll/AutoGPTQ/releases/download/v0.6.0/auto_gptq-0.6.0+cu121-cp310-cp310-linux_x86_64.whl; platform_system == "Linux" and platform_machine == "x86_64" and python_version == "3.10"
 https://github.com/jllllll/AutoGPTQ/releases/download/v0.6.0/auto_gptq-0.6.0+cu121-cp39-cp39-linux_x86_64.whl; platform_system == "Linux" and platform_machine == "x86_64" and python_version == "3.9"
 https://github.com/jllllll/AutoGPTQ/releases/download/v0.6.0/auto_gptq-0.6.0+cu121-cp38-cp38-linux_x86_64.whl; platform_system == "Linux" and platform_machine == "x86_64" and python_version == "3.8"
-https://github.com/turboderp/exllamav2/releases/download/v0.0.11/exllamav2-0.0.11+cu121-cp311-cp311-win_amd64.whl; platform_system == "Windows" and python_version == "3.11"
+https://github.com/turboderp/exllamav2/releases/download/v0.0.12/exllamav2-0.0.12+cu121-cp311-cp311-win_amd64.whl; platform_system == "Windows" and python_version == "3.11"
-https://github.com/turboderp/exllamav2/releases/download/v0.0.11/exllamav2-0.0.11+cu121-cp310-cp310-win_amd64.whl; platform_system == "Windows" and python_version == "3.10"
+https://github.com/turboderp/exllamav2/releases/download/v0.0.12/exllamav2-0.0.12+cu121-cp310-cp310-win_amd64.whl; platform_system == "Windows" and python_version == "3.10"
-https://github.com/turboderp/exllamav2/releases/download/v0.0.11/exllamav2-0.0.11+cu121-cp39-cp39-win_amd64.whl; platform_system == "Windows" and python_version == "3.9"
+https://github.com/turboderp/exllamav2/releases/download/v0.0.12/exllamav2-0.0.12+cu121-cp39-cp39-win_amd64.whl; platform_system == "Windows" and python_version == "3.9"
-https://github.com/turboderp/exllamav2/releases/download/v0.0.11/exllamav2-0.0.11+cu121-cp38-cp38-win_amd64.whl; platform_system == "Windows" and python_version == "3.8"
+https://github.com/turboderp/exllamav2/releases/download/v0.0.12/exllamav2-0.0.12+cu121-cp38-cp38-win_amd64.whl; platform_system == "Windows" and python_version == "3.8"
-https://github.com/turboderp/exllamav2/releases/download/v0.0.11/exllamav2-0.0.11+cu121-cp311-cp311-linux_x86_64.whl; platform_system == "Linux" and platform_machine == "x86_64" and python_version == "3.11"
+https://github.com/turboderp/exllamav2/releases/download/v0.0.12/exllamav2-0.0.12+cu121-cp311-cp311-linux_x86_64.whl; platform_system == "Linux" and platform_machine == "x86_64" and python_version == "3.11"
-https://github.com/turboderp/exllamav2/releases/download/v0.0.11/exllamav2-0.0.11+cu121-cp310-cp310-linux_x86_64.whl; platform_system == "Linux" and platform_machine == "x86_64" and python_version == "3.10"
+https://github.com/turboderp/exllamav2/releases/download/v0.0.12/exllamav2-0.0.12+cu121-cp310-cp310-linux_x86_64.whl; platform_system == "Linux" and platform_machine == "x86_64" and python_version == "3.10"
-https://github.com/turboderp/exllamav2/releases/download/v0.0.11/exllamav2-0.0.11+cu121-cp39-cp39-linux_x86_64.whl; platform_system == "Linux" and platform_machine == "x86_64" and python_version == "3.9"
+https://github.com/turboderp/exllamav2/releases/download/v0.0.12/exllamav2-0.0.12+cu121-cp39-cp39-linux_x86_64.whl; platform_system == "Linux" and platform_machine == "x86_64" and python_version == "3.9"
-https://github.com/turboderp/exllamav2/releases/download/v0.0.11/exllamav2-0.0.11+cu121-cp38-cp38-linux_x86_64.whl; platform_system == "Linux" and platform_machine == "x86_64" and python_version == "3.8"
+https://github.com/turboderp/exllamav2/releases/download/v0.0.12/exllamav2-0.0.12+cu121-cp38-cp38-linux_x86_64.whl; platform_system == "Linux" and platform_machine == "x86_64" and python_version == "3.8"
 https://github.com/jllllll/flash-attention/releases/download/v2.3.4/flash_attn-2.3.4+cu121torch2.1cxx11abiFALSE-cp311-cp311-win_amd64.whl; platform_system == "Windows" and python_version == "3.11"
 https://github.com/jllllll/flash-attention/releases/download/v2.3.4/flash_attn-2.3.4+cu121torch2.1cxx11abiFALSE-cp310-cp310-win_amd64.whl; platform_system == "Windows" and python_version == "3.10"
 https://github.com/jllllll/flash-attention/releases/download/v2.3.4/flash_attn-2.3.4+cu121torch2.1cxx11abiFALSE-cp39-cp39-win_amd64.whl; platform_system == "Windows" and python_version == "3.9"

requirements_amd.txt

@@ -2,9 +2,9 @@ accelerate==0.25.*
 colorama
 datasets
 einops
-exllamav2==0.0.11; platform_system == "Windows" or python_version < "3.10" or python_version > "3.11" or platform_machine != "x86_64"
+exllamav2==0.0.12; platform_system == "Windows" or python_version < "3.10" or python_version > "3.11" or platform_machine != "x86_64"
 gradio==3.50.*
-hqq==0.1.2
+hqq==0.1.2.post1
 jinja2==3.1.2
 lm_eval==0.3.0
 markdown
@@ -47,8 +47,8 @@ https://github.com/jllllll/AutoGPTQ/releases/download/v0.6.0/auto_gptq-0.6.0+roc
 https://github.com/jllllll/AutoGPTQ/releases/download/v0.6.0/auto_gptq-0.6.0+rocm5.6-cp310-cp310-linux_x86_64.whl; platform_system == "Linux" and platform_machine == "x86_64" and python_version == "3.10"
 https://github.com/jllllll/AutoGPTQ/releases/download/v0.6.0/auto_gptq-0.6.0+rocm5.6-cp39-cp39-linux_x86_64.whl; platform_system == "Linux" and platform_machine == "x86_64" and python_version == "3.9"
 https://github.com/jllllll/AutoGPTQ/releases/download/v0.6.0/auto_gptq-0.6.0+rocm5.6-cp38-cp38-linux_x86_64.whl; platform_system == "Linux" and platform_machine == "x86_64" and python_version == "3.8"
-https://github.com/turboderp/exllamav2/releases/download/v0.0.11/exllamav2-0.0.11+rocm5.6-cp311-cp311-linux_x86_64.whl; platform_system == "Linux" and platform_machine == "x86_64" and python_version == "3.11"
+https://github.com/turboderp/exllamav2/releases/download/v0.0.12/exllamav2-0.0.12+rocm5.6-cp311-cp311-linux_x86_64.whl; platform_system == "Linux" and platform_machine == "x86_64" and python_version == "3.11"
-https://github.com/turboderp/exllamav2/releases/download/v0.0.11/exllamav2-0.0.11+rocm5.6-cp310-cp310-linux_x86_64.whl; platform_system == "Linux" and platform_machine == "x86_64" and python_version == "3.10"
+https://github.com/turboderp/exllamav2/releases/download/v0.0.12/exllamav2-0.0.12+rocm5.6-cp310-cp310-linux_x86_64.whl; platform_system == "Linux" and platform_machine == "x86_64" and python_version == "3.10"
 https://github.com/jllllll/GPTQ-for-LLaMa-CUDA/releases/download/0.1.1/gptq_for_llama-0.1.1+rocm5.6-cp311-cp311-linux_x86_64.whl; platform_system == "Linux" and platform_machine == "x86_64" and python_version == "3.11"
 https://github.com/jllllll/GPTQ-for-LLaMa-CUDA/releases/download/0.1.1/gptq_for_llama-0.1.1+rocm5.6-cp310-cp310-linux_x86_64.whl; platform_system == "Linux" and platform_machine == "x86_64" and python_version == "3.10"
 https://github.com/jllllll/GPTQ-for-LLaMa-CUDA/releases/download/0.1.1/gptq_for_llama-0.1.1+rocm5.6-cp39-cp39-linux_x86_64.whl; platform_system == "Linux" and platform_machine == "x86_64" and python_version == "3.9"

requirements_amd_noavx2.txt

@@ -2,9 +2,9 @@ accelerate==0.25.*
 colorama
 datasets
 einops
-exllamav2==0.0.11; platform_system == "Windows" or python_version < "3.10" or python_version > "3.11" or platform_machine != "x86_64"
+exllamav2==0.0.12; platform_system == "Windows" or python_version < "3.10" or python_version > "3.11" or platform_machine != "x86_64"
 gradio==3.50.*
-hqq==0.1.2
+hqq==0.1.2.post1
 jinja2==3.1.2
 lm_eval==0.3.0
 markdown
@@ -43,8 +43,8 @@ https://github.com/jllllll/AutoGPTQ/releases/download/v0.6.0/auto_gptq-0.6.0+roc
 https://github.com/jllllll/AutoGPTQ/releases/download/v0.6.0/auto_gptq-0.6.0+rocm5.6-cp310-cp310-linux_x86_64.whl; platform_system == "Linux" and platform_machine == "x86_64" and python_version == "3.10"
 https://github.com/jllllll/AutoGPTQ/releases/download/v0.6.0/auto_gptq-0.6.0+rocm5.6-cp39-cp39-linux_x86_64.whl; platform_system == "Linux" and platform_machine == "x86_64" and python_version == "3.9"
 https://github.com/jllllll/AutoGPTQ/releases/download/v0.6.0/auto_gptq-0.6.0+rocm5.6-cp38-cp38-linux_x86_64.whl; platform_system == "Linux" and platform_machine == "x86_64" and python_version == "3.8"
-https://github.com/turboderp/exllamav2/releases/download/v0.0.11/exllamav2-0.0.11+rocm5.6-cp311-cp311-linux_x86_64.whl; platform_system == "Linux" and platform_machine == "x86_64" and python_version == "3.11"
+https://github.com/turboderp/exllamav2/releases/download/v0.0.12/exllamav2-0.0.12+rocm5.6-cp311-cp311-linux_x86_64.whl; platform_system == "Linux" and platform_machine == "x86_64" and python_version == "3.11"
-https://github.com/turboderp/exllamav2/releases/download/v0.0.11/exllamav2-0.0.11+rocm5.6-cp310-cp310-linux_x86_64.whl; platform_system == "Linux" and platform_machine == "x86_64" and python_version == "3.10"
+https://github.com/turboderp/exllamav2/releases/download/v0.0.12/exllamav2-0.0.12+rocm5.6-cp310-cp310-linux_x86_64.whl; platform_system == "Linux" and platform_machine == "x86_64" and python_version == "3.10"
 https://github.com/jllllll/GPTQ-for-LLaMa-CUDA/releases/download/0.1.1/gptq_for_llama-0.1.1+rocm5.6-cp311-cp311-linux_x86_64.whl; platform_system == "Linux" and platform_machine == "x86_64" and python_version == "3.11"
 https://github.com/jllllll/GPTQ-for-LLaMa-CUDA/releases/download/0.1.1/gptq_for_llama-0.1.1+rocm5.6-cp310-cp310-linux_x86_64.whl; platform_system == "Linux" and platform_machine == "x86_64" and python_version == "3.10"
 https://github.com/jllllll/GPTQ-for-LLaMa-CUDA/releases/download/0.1.1/gptq_for_llama-0.1.1+rocm5.6-cp39-cp39-linux_x86_64.whl; platform_system == "Linux" and platform_machine == "x86_64" and python_version == "3.9"

requirements_apple_intel.txt

@@ -2,9 +2,9 @@ accelerate==0.25.*
 colorama
 datasets
 einops
-exllamav2==0.0.11
+exllamav2==0.0.12
 gradio==3.50.*
-hqq==0.1.2
+hqq==0.1.2.post1
 jinja2==3.1.2
 lm_eval==0.3.0
 markdown

requirements_apple_silicon.txt

@@ -2,9 +2,9 @@ accelerate==0.25.*
 colorama
 datasets
 einops
-exllamav2==0.0.11
+exllamav2==0.0.12
 gradio==3.50.*
-hqq==0.1.2
+hqq==0.1.2.post1
 jinja2==3.1.2
 lm_eval==0.3.0
 markdown

requirements_cpu_only.txt

@@ -2,9 +2,9 @@ accelerate==0.25.*
 colorama
 datasets
 einops
-exllamav2==0.0.11
+exllamav2==0.0.12
 gradio==3.50.*
-hqq==0.1.2
+hqq==0.1.2.post1
 jinja2==3.1.2
 lm_eval==0.3.0
 markdown

requirements_cpu_only_noavx2.txt

@@ -2,9 +2,9 @@ accelerate==0.25.*
 colorama
 datasets
 einops
-exllamav2==0.0.11
+exllamav2==0.0.12
 gradio==3.50.*
-hqq==0.1.2
+hqq==0.1.2.post1
 jinja2==3.1.2
 lm_eval==0.3.0
 markdown

requirements_noavx2.txt

@@ -2,9 +2,9 @@ accelerate==0.25.*
 colorama
 datasets
 einops
-exllamav2==0.0.11; platform_system != "Darwin" and platform_machine != "x86_64"
+exllamav2==0.0.12; platform_system != "Darwin" and platform_machine != "x86_64"
 gradio==3.50.*
-hqq==0.1.2
+hqq==0.1.2.post1
 jinja2==3.1.2
 lm_eval==0.3.0
 markdown
@@ -67,14 +67,14 @@ https://github.com/jllllll/AutoGPTQ/releases/download/v0.6.0/auto_gptq-0.6.0+cu1
 https://github.com/jllllll/AutoGPTQ/releases/download/v0.6.0/auto_gptq-0.6.0+cu121-cp310-cp310-linux_x86_64.whl; platform_system == "Linux" and platform_machine == "x86_64" and python_version == "3.10"
 https://github.com/jllllll/AutoGPTQ/releases/download/v0.6.0/auto_gptq-0.6.0+cu121-cp39-cp39-linux_x86_64.whl; platform_system == "Linux" and platform_machine == "x86_64" and python_version == "3.9"
 https://github.com/jllllll/AutoGPTQ/releases/download/v0.6.0/auto_gptq-0.6.0+cu121-cp38-cp38-linux_x86_64.whl; platform_system == "Linux" and platform_machine == "x86_64" and python_version == "3.8"
-https://github.com/turboderp/exllamav2/releases/download/v0.0.11/exllamav2-0.0.11+cu121-cp311-cp311-win_amd64.whl; platform_system == "Windows" and python_version == "3.11"
+https://github.com/turboderp/exllamav2/releases/download/v0.0.12/exllamav2-0.0.12+cu121-cp311-cp311-win_amd64.whl; platform_system == "Windows" and python_version == "3.11"
-https://github.com/turboderp/exllamav2/releases/download/v0.0.11/exllamav2-0.0.11+cu121-cp310-cp310-win_amd64.whl; platform_system == "Windows" and python_version == "3.10"
+https://github.com/turboderp/exllamav2/releases/download/v0.0.12/exllamav2-0.0.12+cu121-cp310-cp310-win_amd64.whl; platform_system == "Windows" and python_version == "3.10"
-https://github.com/turboderp/exllamav2/releases/download/v0.0.11/exllamav2-0.0.11+cu121-cp39-cp39-win_amd64.whl; platform_system == "Windows" and python_version == "3.9"
+https://github.com/turboderp/exllamav2/releases/download/v0.0.12/exllamav2-0.0.12+cu121-cp39-cp39-win_amd64.whl; platform_system == "Windows" and python_version == "3.9"
-https://github.com/turboderp/exllamav2/releases/download/v0.0.11/exllamav2-0.0.11+cu121-cp38-cp38-win_amd64.whl; platform_system == "Windows" and python_version == "3.8"
+https://github.com/turboderp/exllamav2/releases/download/v0.0.12/exllamav2-0.0.12+cu121-cp38-cp38-win_amd64.whl; platform_system == "Windows" and python_version == "3.8"
-https://github.com/turboderp/exllamav2/releases/download/v0.0.11/exllamav2-0.0.11+cu121-cp311-cp311-linux_x86_64.whl; platform_system == "Linux" and platform_machine == "x86_64" and python_version == "3.11"
+https://github.com/turboderp/exllamav2/releases/download/v0.0.12/exllamav2-0.0.12+cu121-cp311-cp311-linux_x86_64.whl; platform_system == "Linux" and platform_machine == "x86_64" and python_version == "3.11"
-https://github.com/turboderp/exllamav2/releases/download/v0.0.11/exllamav2-0.0.11+cu121-cp310-cp310-linux_x86_64.whl; platform_system == "Linux" and platform_machine == "x86_64" and python_version == "3.10"
+https://github.com/turboderp/exllamav2/releases/download/v0.0.12/exllamav2-0.0.12+cu121-cp310-cp310-linux_x86_64.whl; platform_system == "Linux" and platform_machine == "x86_64" and python_version == "3.10"
-https://github.com/turboderp/exllamav2/releases/download/v0.0.11/exllamav2-0.0.11+cu121-cp39-cp39-linux_x86_64.whl; platform_system == "Linux" and platform_machine == "x86_64" and python_version == "3.9"
+https://github.com/turboderp/exllamav2/releases/download/v0.0.12/exllamav2-0.0.12+cu121-cp39-cp39-linux_x86_64.whl; platform_system == "Linux" and platform_machine == "x86_64" and python_version == "3.9"
-https://github.com/turboderp/exllamav2/releases/download/v0.0.11/exllamav2-0.0.11+cu121-cp38-cp38-linux_x86_64.whl; platform_system == "Linux" and platform_machine == "x86_64" and python_version == "3.8"
+https://github.com/turboderp/exllamav2/releases/download/v0.0.12/exllamav2-0.0.12+cu121-cp38-cp38-linux_x86_64.whl; platform_system == "Linux" and platform_machine == "x86_64" and python_version == "3.8"
 https://github.com/jllllll/flash-attention/releases/download/v2.3.4/flash_attn-2.3.4+cu121torch2.1cxx11abiFALSE-cp311-cp311-win_amd64.whl; platform_system == "Windows" and python_version == "3.11"
 https://github.com/jllllll/flash-attention/releases/download/v2.3.4/flash_attn-2.3.4+cu121torch2.1cxx11abiFALSE-cp310-cp310-win_amd64.whl; platform_system == "Windows" and python_version == "3.10"
 https://github.com/jllllll/flash-attention/releases/download/v2.3.4/flash_attn-2.3.4+cu121torch2.1cxx11abiFALSE-cp39-cp39-win_amd64.whl; platform_system == "Windows" and python_version == "3.9"

requirements_nowheels.txt

@@ -2,9 +2,9 @@ accelerate==0.25.*
 colorama
 datasets
 einops
-exllamav2==0.0.11
+exllamav2==0.0.12
 gradio==3.50.*
-hqq==0.1.2
+hqq==0.1.2.post1
 jinja2==3.1.2
 lm_eval==0.3.0
 markdown