diff --git a/.gitignore b/.gitignore
index ca307c4a..318e147d 100644
--- a/.gitignore
+++ b/.gitignore
@@ -26,6 +26,7 @@
 .DS_Store
 .eslintrc.js
 .idea
+.installer_state.json
 .venv
 venv
 .envrc
diff --git a/README.md b/README.md
index f1633334..3642fc58 100644
--- a/README.md
+++ b/README.md
@@ -380,7 +380,7 @@ text-generation-webui
 │   │   └── tokenizer.model
 ```
 
-In both cases, you can use the "Model" tab of the UI to download the model from Hugging Face automatically. It is also possible to download it via the command-line with 
+In both cases, you can use the "Model" tab of the UI to download the model from Hugging Face automatically. It is also possible to download it via the command-line with:
 
 ```
 python download-model.py organization/model
diff --git a/css/main.css b/css/main.css
index 1a7efe70..b10d1980 100644
--- a/css/main.css
+++ b/css/main.css
@@ -1259,6 +1259,16 @@ div.svelte-362y77>*, div.svelte-362y77>.form>* {
     left: 25px;
 }
 
+.footer-button.footer-continue-button {
+    bottom: -23px;
+    left: 50px;
+}
+
+.footer-button.footer-remove-button {
+    bottom: -23px;
+    left: 75px;
+}
+
 .message:hover .footer-button,
 .user-message:hover .footer-button,
 .assistant-message:hover .footer-button {
diff --git a/docs/12 - OpenAI API.md b/docs/12 - OpenAI API.md
index 9b4f89bf..364c6b09 100644
--- a/docs/12 - OpenAI API.md	
+++ b/docs/12 - OpenAI API.md	
@@ -14,7 +14,7 @@ Add `--api` to your command-line flags.
 * To create a public Cloudflare URL, add the `--public-api` flag.
 * To listen on your local network, add the `--listen` flag.
 * To change the port, which is 5000 by default, use `--api-port 1234` (change 1234 to your desired port number).
-* To use SSL, add `--ssl-keyfile key.pem --ssl-certfile cert.pem`. Note that it doesn't work with `--public-api`.
+* To use SSL, add `--ssl-keyfile key.pem --ssl-certfile cert.pem`. ⚠️ **Note**: this doesn't work with `--public-api` since Cloudflare already uses HTTPS by default.
 * To use an API key for authentication, add `--api-key yourkey`.
 
 ### Examples
@@ -51,8 +51,7 @@ curl http://127.0.0.1:5000/v1/chat/completions \
         "content": "Hello!"
       }
     ],
-    "mode": "instruct",
-    "instruction_template": "Alpaca"
+    "mode": "instruct"
   }'
 ```
 
@@ -86,7 +85,6 @@ curl http://127.0.0.1:5000/v1/chat/completions \
       }
     ],
     "mode": "instruct",
-    "instruction_template": "Alpaca",
     "stream": true
   }'
 ```
@@ -131,9 +129,6 @@ curl -k http://127.0.0.1:5000/v1/internal/model/load \
     "args": {
       "load_in_4bit": true,
       "n_gpu_layers": 12
-    },
-    "settings": {
-      "instruction_template": "Alpaca"
     }
   }'
 ```
@@ -198,7 +193,7 @@ while True:
     assistant_message = ''
     for event in client.events():
         payload = json.loads(event.data)
-        chunk = payload['choices'][0]['message']['content']
+        chunk = payload['choices'][0]['delta']['content']
         assistant_message += chunk
         print(chunk, end='')
 
@@ -241,6 +236,27 @@ for event in client.events():
 print()
 ```
 
+#### Python example with API key
+
+Replace
+
+```python
+headers = {
+    "Content-Type": "application/json"
+}
+```
+
+with
+
+```python
+headers = {
+    "Content-Type": "application/json",
+    "Authorization": "Bearer yourPassword123"
+}
+```
+
+in any of the examples above.
+
 ### Environment variables
 
 The following environment variables can be used (they take precedence over everything else):
diff --git a/download-model.py b/download-model.py
index 8fe94371..8ff1d69c 100644
--- a/download-model.py
+++ b/download-model.py
@@ -14,6 +14,7 @@ import json
 import os
 import re
 import sys
+from multiprocessing import Array
 from pathlib import Path
 from time import sleep
 
@@ -27,9 +28,10 @@ base = os.environ.get("HF_ENDPOINT") or "https://huggingface.co"
 
 
 class ModelDownloader:
-    def __init__(self, max_retries=5):
+    def __init__(self, max_retries=7):
         self.max_retries = max_retries
         self.session = self.get_session()
+        self._progress_bar_slots = None
 
     def get_session(self):
         session = requests.Session()
@@ -186,73 +188,112 @@ class ModelDownloader:
         output_folder = Path(base_folder) / output_folder
         return output_folder
 
+    @property
+    def progress_bar_slots(self):
+        if self._progress_bar_slots is None:
+            raise RuntimeError("Progress bar slots not initialized. Start download threads first.")
+
+        return self._progress_bar_slots
+
+    def initialize_progress_bar_slots(self, num_threads):
+        self._progress_bar_slots = Array("B", [0] * num_threads)
+
+    def get_progress_bar_position(self):
+        with self.progress_bar_slots.get_lock():
+            for i in range(len(self.progress_bar_slots)):
+                if self.progress_bar_slots[i] == 0:
+                    self.progress_bar_slots[i] = 1
+                    return i
+
+        return 0  # fallback
+
+    def release_progress_bar_position(self, slot):
+        with self.progress_bar_slots.get_lock():
+            self.progress_bar_slots[slot] = 0
+
     def get_single_file(self, url, output_folder, start_from_scratch=False):
         filename = Path(url.rsplit('/', 1)[1])
         output_path = output_folder / filename
+        progress_bar_position = self.get_progress_bar_position()
 
-        max_retries = 7
+        max_retries = self.max_retries
         attempt = 0
-        while attempt < max_retries:
-            attempt += 1
-            session = self.session
-            headers = {}
-            mode = 'wb'
+        try:
+            while attempt < max_retries:
+                attempt += 1
+                session = self.session
+                headers = {}
+                mode = 'wb'
 
-            try:
-                if output_path.exists() and not start_from_scratch:
-                    # Resume download
-                    r = session.get(url, stream=True, timeout=20)
-                    total_size = int(r.headers.get('content-length', 0))
-                    if output_path.stat().st_size >= total_size:
-                        return
+                try:
+                    if output_path.exists() and not start_from_scratch:
+                        # Resume download
+                        r = session.get(url, stream=True, timeout=20)
+                        total_size = int(r.headers.get('content-length', 0))
+                        if output_path.stat().st_size >= total_size:
+                            return
 
-                    headers = {'Range': f'bytes={output_path.stat().st_size}-'}
-                    mode = 'ab'
+                        headers = {'Range': f'bytes={output_path.stat().st_size}-'}
+                        mode = 'ab'
 
-                with session.get(url, stream=True, headers=headers, timeout=30) as r:
-                    r.raise_for_status()  # If status is not 2xx, raise an error
-                    total_size = int(r.headers.get('content-length', 0))
-                    block_size = 1024 * 1024  # 1MB
+                    with session.get(url, stream=True, headers=headers, timeout=30) as r:
+                        r.raise_for_status()  # If status is not 2xx, raise an error
+                        total_size = int(r.headers.get('content-length', 0))
+                        block_size = 1024 * 1024  # 1MB
 
-                    filename_str = str(filename)  # Convert PosixPath to string if necessary
+                        filename_str = str(filename)  # Convert PosixPath to string if necessary
 
-                    tqdm_kwargs = {
-                        'total': total_size,
-                        'unit': 'B',
-                        'unit_scale': True,
-                        'unit_divisor': 1024,
-                        'bar_format': '{desc}{percentage:3.0f}%|{bar:50}| {n_fmt}/{total_fmt} [{elapsed}<{remaining}, {rate_fmt}]',
-                        'desc': f"{filename_str}: "
-                    }
+                        tqdm_kwargs = {
+                            'total': total_size,
+                            'unit': 'B',
+                            'unit_scale': True,
+                            'unit_divisor': 1024,
+                            'bar_format': '{desc}{percentage:3.0f}%|{bar:50}| {n_fmt}/{total_fmt} [{elapsed}<{remaining}, {rate_fmt}]',
+                            'desc': f"{filename_str}: ",
+                            'position': progress_bar_position,
+                            'leave': False
+                        }
 
-                    if 'COLAB_GPU' in os.environ:
-                        tqdm_kwargs.update({
-                            'position': 0,
-                            'leave': True
-                        })
+                        if 'COLAB_GPU' in os.environ:
+                            tqdm_kwargs.update({
+                                'position': 0,
+                                'leave': True
+                            })
 
-                    with open(output_path, mode) as f:
-                        with tqdm.tqdm(**tqdm_kwargs) as t:
-                            count = 0
-                            for data in r.iter_content(block_size):
-                                f.write(data)
-                                t.update(len(data))
-                                if total_size != 0 and self.progress_bar is not None:
-                                    count += len(data)
-                                    self.progress_bar(float(count) / float(total_size), f"{filename_str}")
+                        with open(output_path, mode) as f:
+                            with tqdm.tqdm(**tqdm_kwargs) as t:
+                                count = 0
+                                for data in r.iter_content(block_size):
+                                    f.write(data)
+                                    t.update(len(data))
+                                    if total_size != 0 and self.progress_bar is not None:
+                                        count += len(data)
+                                        self.progress_bar(float(count) / float(total_size), f"{filename_str}")
 
-                    break  # Exit loop if successful
-            except (RequestException, ConnectionError, Timeout) as e:
-                print(f"Error downloading {filename}: {e}.")
-                print(f"That was attempt {attempt}/{max_retries}.", end=' ')
-                if attempt < max_retries:
-                    print(f"Retry begins in {2 ** attempt} seconds.")
-                    sleep(2 ** attempt)
-                else:
-                    print("Failed to download after the maximum number of attempts.")
+                        break  # Exit loop if successful
+                except (RequestException, ConnectionError, Timeout) as e:
+                    print(f"Error downloading {filename}: {e}.")
+                    print(f"That was attempt {attempt}/{max_retries}.", end=' ')
+                    if attempt < max_retries:
+                        print(f"Retry begins in {2 ** attempt} seconds.")
+                        sleep(2 ** attempt)
+                    else:
+                        print("Failed to download after the maximum number of attempts.")
+        finally:
+            self.release_progress_bar_position(progress_bar_position)
 
     def start_download_threads(self, file_list, output_folder, start_from_scratch=False, threads=4):
-        thread_map(lambda url: self.get_single_file(url, output_folder, start_from_scratch=start_from_scratch), file_list, max_workers=threads, disable=True)
+        self.initialize_progress_bar_slots(threads)
+        tqdm.tqdm.set_lock(tqdm.tqdm.get_lock())
+        try:
+            thread_map(
+                lambda url: self.get_single_file(url, output_folder, start_from_scratch=start_from_scratch),
+                file_list,
+                max_workers=threads,
+                disable=True
+            )
+        finally:
+            print(f"\nDownload of {len(file_list)} files to {output_folder} completed.")
 
     def download_model_files(self, model, branch, links, sha256, output_folder, progress_bar=None, start_from_scratch=False, threads=4, specific_file=None, is_llamacpp=False):
         self.progress_bar = progress_bar
@@ -318,7 +359,7 @@ if __name__ == '__main__':
     parser.add_argument('--model-dir', type=str, default=None, help='Save the model files to a subfolder of this folder instead of the default one (text-generation-webui/models).')
     parser.add_argument('--clean', action='store_true', help='Does not resume the previous download.')
     parser.add_argument('--check', action='store_true', help='Validates the checksums of model files.')
-    parser.add_argument('--max-retries', type=int, default=5, help='Max retries count when get error in download time.')
+    parser.add_argument('--max-retries', type=int, default=7, help='Max retries count when get error in download time.')
     args = parser.parse_args()
 
     branch = args.branch
diff --git a/extensions/Training_PRO/script.py b/extensions/Training_PRO/script.py
index 01bcf67d..f553e482 100644
--- a/extensions/Training_PRO/script.py
+++ b/extensions/Training_PRO/script.py
@@ -557,12 +557,6 @@ def calc_trainable_parameters(model):
 
 def do_train(lora_name: str, always_override: bool, save_steps: int, micro_batch_size: int, batch_size: int, epochs: int, learning_rate: str, lr_scheduler_type: str, lora_rank: int, lora_alpha: int, lora_dropout: float, cutoff_len: int, dataset: str, eval_dataset: str, format: str, eval_steps: int, raw_text_file: str, higher_rank_limit: bool, warmup_steps: int, optimizer: str, hard_cut_string: str, train_only_after: str, stop_at_loss: float, add_eos_token: bool, min_chars: int, report_to: str, precize_slicing_overlap: bool, add_eos_token_type: str, save_steps_under_loss: float, add_bos_token: bool, training_projection: str,sliding_window:bool,warmup_ratio:float, grad_accumulation: int,neft_noise_alpha:float):
 
-    if shared.args.monkey_patch:
-        from alpaca_lora_4bit.monkeypatch.peft_tuners_lora_monkey_patch import (
-            replace_peft_model_with_int4_lora_model
-        )
-        replace_peft_model_with_int4_lora_model()
-    
     global train_log_graph
     global WANT_INTERRUPT
     WANT_INTERRUPT = False
@@ -600,10 +594,6 @@ def do_train(lora_name: str, always_override: bool, save_steps: int, micro_batch
 
         time.sleep(5)
 
-    if shared.args.loader == 'GPTQ-for-LLaMa' and not shared.args.monkey_patch:
-        yield "LoRA training with GPTQ-for-LLaMa requires loading with `--monkey-patch`", zero_pd
-        return
-
     if cutoff_len <= 0 or micro_batch_size <= 0 or actual_lr <= 0 or lora_rank <= 0 or lora_alpha <= 0:
         yield "Cannot input zeroes.", zero_pd
         return
@@ -865,15 +855,6 @@ def do_train(lora_name: str, always_override: bool, save_steps: int, micro_batch
         yield traceback.format_exc().replace('\n', '\n\n'), zero_pd
         return
 
-    if shared.args.monkey_patch:
-        from alpaca_lora_4bit.autograd_4bit import Autograd4bitQuantLinear
-        from alpaca_lora_4bit.models import Linear4bitLt
-        for _, m in lora_model.named_modules():
-            if isinstance(m, Autograd4bitQuantLinear) or isinstance(m, Linear4bitLt):
-                if m.is_v1_model:
-                    m.zeros = m.zeros.half()
-                m.scales = m.scales.half()
-
     class Tracked():
         def __init__(self):
             self.current_steps = 0
diff --git a/extensions/openai/completions.py b/extensions/openai/completions.py
index 2cefc22b..f1a60645 100644
--- a/extensions/openai/completions.py
+++ b/extensions/openai/completions.py
@@ -146,7 +146,7 @@ def convert_history(history):
                 for item in entry['content']:
                     if not isinstance(item, dict):
                         continue
-                    
+
                     image_url = None
                     content = None
                     if item['type'] == 'image_url' and isinstance(item['image_url'], dict):
@@ -205,7 +205,7 @@ def convert_history(history):
             else:
                 chat_dialogue.append(['', current_reply])
         elif role == "system":
-            system_message = content
+            system_message += f"\n{content}" if system_message else content
 
     if not user_input_last:
         user_input = ""
diff --git a/js/global_scope_js.js b/js/global_scope_js.js
index 6bf0f0e3..f308edb9 100644
--- a/js/global_scope_js.js
+++ b/js/global_scope_js.js
@@ -22,6 +22,14 @@ function regenerateClick() {
   document.getElementById("Regenerate").click();
 }
 
+function continueClick() {
+  document.getElementById("Continue").click();
+}
+
+function removeLastClick() {
+  document.getElementById("Remove-last").click();
+}
+
 function handleMorphdomUpdate(text) {
   morphdom(
     document.getElementById("chat").parentNode,
diff --git a/modules/chat.py b/modules/chat.py
index 694c137b..0e47da29 100644
--- a/modules/chat.py
+++ b/modules/chat.py
@@ -30,8 +30,13 @@ from modules.text_generation import (
 )
 from modules.utils import delete_file, get_available_characters, save_file
 
-# Copied from the Transformers library
+
+def strftime_now(format):
+    return datetime.now().strftime(format)
+
+
 jinja_env = ImmutableSandboxedEnvironment(trim_blocks=True, lstrip_blocks=True)
+jinja_env.globals["strftime_now"] = strftime_now
 
 
 def str_presenter(dumper, data):
diff --git a/modules/html_generator.py b/modules/html_generator.py
index 29973412..3edbef5e 100644
--- a/modules/html_generator.py
+++ b/modules/html_generator.py
@@ -154,6 +154,8 @@ def add_long_list_class(html):
 
 @functools.lru_cache(maxsize=None)
 def convert_to_markdown(string):
+    if not string:
+        return ""
 
     # Make \[ \]  LaTeX equations inline
     pattern = r'^\s*\\\[\s*\n([\s\S]*?)\n\s*\\\]\s*$'
@@ -304,8 +306,13 @@ def get_image_cache(path):
 
 copy_svg = '''<svg xmlns="http://www.w3.org/2000/svg" width="20" height="20" viewBox="0 0 24 24" fill="none" stroke="currentColor" stroke-width="2" stroke-linecap="round" stroke-linejoin="round" class="tabler-icon tabler-icon-copy"><path d="M8 8m0 2a2 2 0 0 1 2 -2h8a2 2 0 0 1 2 2v8a2 2 0 0 1 -2 2h-8a2 2 0 0 1 -2 -2z"></path><path d="M16 8v-2a2 2 0 0 0 -2 -2h-8a2 2 0 0 0 -2 2v8a2 2 0 0 0 2 2h2"></path></svg>'''
 refresh_svg = '''<svg xmlns="http://www.w3.org/2000/svg" width="20" height="20" viewBox="0 0 24 24" fill="none" stroke="currentColor" stroke-width="2" stroke-linecap="round" stroke-linejoin="round" class="tabler-icon tabler-icon-repeat"><path d="M4 12v-3a3 3 0 0 1 3 -3h13m-3 -3l3 3l-3 3"></path><path d="M20 12v3a3 3 0 0 1 -3 3h-13m3 3l-3 -3l3 -3"></path></svg>'''
-copy_button = f'<button class="footer-button footer-copy-button" onclick="copyToClipboard(this)">{copy_svg}</button>'
-refresh_button = f'<button class="footer-button footer-refresh-button" onclick="regenerateClick()">{refresh_svg}</button>'
+continue_svg = '''<svg  xmlns="http://www.w3.org/2000/svg"  width="20"  height="20"  viewBox="0 0 24 24"  fill="none"  stroke="currentColor"  stroke-width="2"  stroke-linecap="round"  stroke-linejoin="round"  class="icon icon-tabler icons-tabler-outline icon-tabler-player-play"><path stroke="none" d="M0 0h24v24H0z" fill="none"/><path d="M7 4v16l13 -8z" /></svg>'''
+remove_svg = '''<svg  xmlns="http://www.w3.org/2000/svg"  width="20"  height="20"  viewBox="0 0 24 24"  fill="none"  stroke="currentColor"  stroke-width="2"  stroke-linecap="round"  stroke-linejoin="round"  class="icon icon-tabler icons-tabler-outline icon-tabler-trash"><path stroke="none" d="M0 0h24v24H0z" fill="none"/><path d="M4 7l16 0" /><path d="M10 11l0 6" /><path d="M14 11l0 6" /><path d="M5 7l1 12a2 2 0 0 0 2 2h8a2 2 0 0 0 2 -2l1 -12" /><path d="M9 7v-3a1 1 0 0 1 1 -1h4a1 1 0 0 1 1 1v3" /></svg>'''
+
+copy_button = f'<button class="footer-button footer-copy-button" title="Copy" onclick="copyToClipboard(this)">{copy_svg}</button>'
+refresh_button = f'<button class="footer-button footer-refresh-button" title="Regenerate" onclick="regenerateClick()">{refresh_svg}</button>'
+continue_button = f'<button class="footer-button footer-continue-button" title="Continue" onclick="continueClick()">{continue_svg}</button>'
+remove_button = f'<button class="footer-button footer-remove-button" title="Remove last reply" onclick="removeLastClick()">{remove_svg}</button>'
 
 
 def generate_instruct_html(history):
@@ -334,6 +341,8 @@ def generate_instruct_html(history):
             f'<div class="message-body">{converted_visible[1]}</div>'
             f'{copy_button}'
             f'{refresh_button if i == len(history["visible"]) - 1 else ""}'
+            f'{continue_button if i == len(history["visible"]) - 1 else ""}'
+            f'{remove_button if i == len(history["visible"]) - 1 else ""}'
             f'</div>'
             f'</div>'
         )
@@ -383,6 +392,8 @@ def generate_cai_chat_html(history, name1, name2, style, character, reset_cache=
             f'<div class="message-body">{converted_visible[1]}</div>'
             f'{copy_button}'
             f'{refresh_button if i == len(history["visible"]) - 1 else ""}'
+            f'{continue_button if i == len(history["visible"]) - 1 else ""}'
+            f'{remove_button if i == len(history["visible"]) - 1 else ""}'
             f'</div>'
             f'</div>'
         )
@@ -417,6 +428,8 @@ def generate_chat_html(history, name1, name2, reset_cache=False):
             f'<div class="message-body">{converted_visible[1]}</div>'
             f'{copy_button}'
             f'{refresh_button if i == len(history["visible"]) - 1 else ""}'
+            f'{continue_button if i == len(history["visible"]) - 1 else ""}'
+            f'{remove_button if i == len(history["visible"]) - 1 else ""}'
             f'</div>'
             f'</div>'
         )
diff --git a/modules/llamacpp_hf.py b/modules/llamacpp_hf.py
index f9964fe8..b3761e0f 100644
--- a/modules/llamacpp_hf.py
+++ b/modules/llamacpp_hf.py
@@ -202,7 +202,19 @@ class LlamacppHF(PreTrainedModel):
             params["type_v"] = get_llamacpp_cache_type_for_string(shared.args.cache_type)
 
         Llama = llama_cpp_lib().Llama
-        model = Llama(**params)
+        try:
+            model = Llama(**params)
+        except Exception as e:
+            error_message = (
+                f"Failed loading the model. **This usually happens due to lack of memory**. Try these steps:\n"
+                f"1. Reduce the context length `n_ctx` (currently {shared.args.n_ctx})."
+                f"{' Try a lower value like 4096.' if shared.args.n_ctx > 4096 else '.'}"
+                "\n"
+                f"2. Lower the `n-gpu-layers` value (currently {shared.args.n_gpu_layers})."
+            )
+
+            raise type(e)(error_message) from e
+
         model.last_updated_index = -1
 
         return LlamacppHF(model, model_file)
diff --git a/modules/llamacpp_model.py b/modules/llamacpp_model.py
index c79755e4..db25c66c 100644
--- a/modules/llamacpp_model.py
+++ b/modules/llamacpp_model.py
@@ -108,7 +108,19 @@ class LlamaCppModel:
             params["type_k"] = get_llamacpp_cache_type_for_string(shared.args.cache_type)
             params["type_v"] = get_llamacpp_cache_type_for_string(shared.args.cache_type)
 
-        result.model = Llama(**params)
+        try:
+            result.model = Llama(**params)
+        except Exception as e:
+            error_message = (
+                f"Failed loading the model. **This usually happens due to lack of memory**. Try these steps:\n"
+                f"1. Reduce the context length `n_ctx` (currently {shared.args.n_ctx})."
+                f"{' Try a lower value like 4096.' if shared.args.n_ctx > 4096 else '.'}"
+                "\n"
+                f"2. Lower the `n-gpu-layers` value (currently {shared.args.n_gpu_layers})."
+            )
+
+            raise type(e)(error_message) from e
+
         if cache_capacity > 0:
             result.model.set_cache(LlamaCache(capacity_bytes=cache_capacity))
 
diff --git a/modules/ui_model_menu.py b/modules/ui_model_menu.py
index d5116938..1264a9fd 100644
--- a/modules/ui_model_menu.py
+++ b/modules/ui_model_menu.py
@@ -89,8 +89,8 @@ def create_ui():
                             shared.gradio['threads_batch'] = gr.Slider(label="threads_batch", minimum=0, step=1, maximum=256, value=shared.args.threads_batch)
                             shared.gradio['n_batch'] = gr.Slider(label="n_batch", minimum=1, maximum=2048, step=1, value=shared.args.n_batch)
                             shared.gradio['hqq_backend'] = gr.Dropdown(label="hqq_backend", choices=["PYTORCH", "PYTORCH_COMPILE", "ATEN"], value=shared.args.hqq_backend)
-                            shared.gradio['n_ctx'] = gr.Number(label="n_ctx", precision=0, step=256, value=shared.args.n_ctx, info='Context length. ⚠️ Lower this value if you can\'t load the model.')
-                            shared.gradio['max_seq_len'] = gr.Number(label='max_seq_len', precision=0, step=256, value=shared.args.max_seq_len, info='Context length. ⚠️ Lower this value if you can\'t load the model.')
+                            shared.gradio['n_ctx'] = gr.Number(label="n_ctx", precision=0, step=256, value=shared.args.n_ctx, info='Context length. ⚠️ Lower this value if you can\'t load the model. Common values: 2048, 4096, 8192, 16384, 32768.')
+                            shared.gradio['max_seq_len'] = gr.Number(label='max_seq_len', precision=0, step=256, value=shared.args.max_seq_len, info='Context length. ⚠️ Lower this value if you can\'t load the model. Common values: 2048, 4096, 8192, 16384, 32768.')
                             shared.gradio['cache_type'] = gr.Dropdown(label="cache_type", choices=['fp16', 'q8_0', 'q4_0', 'fp8', 'q8', 'q6', 'q4'], value=shared.args.cache_type, info='Valid options: llama.cpp - fp16, q8_0, q4_0; ExLlamaV2 - fp16, fp8, q8, q6, q4.')
                             shared.gradio['tensor_split'] = gr.Textbox(label='tensor_split', info='List of proportions to split the model across multiple GPUs. Example: 60,40')
                             shared.gradio['gpu_split'] = gr.Textbox(label='gpu-split', info='Comma-separated list of VRAM (in GB) to use per GPU. Example: 20,7,7')
diff --git a/one_click.py b/one_click.py
index e78a2450..effc7d43 100644
--- a/one_click.py
+++ b/one_click.py
@@ -1,6 +1,7 @@
 import argparse
 import glob
 import hashlib
+import json
 import os
 import platform
 import re
@@ -101,32 +102,25 @@ def torch_version():
 
 
 def update_pytorch():
-    print_big_message("Checking for PyTorch updates")
-
+    print_big_message("Checking for PyTorch updates.")
     torver = torch_version()
-    is_cuda = '+cu' in torver
-    is_cuda118 = '+cu118' in torver  # 2.1.0+cu118
-    is_rocm = '+rocm' in torver  # 2.0.1+rocm5.4.2
-    is_intel = '+cxx11' in torver  # 2.0.1a0+cxx11.abi
-    is_cpu = '+cpu' in torver  # 2.0.1+cpu
+    base_cmd = f"python -m pip install --upgrade torch=={TORCH_VERSION} torchvision=={TORCHVISION_VERSION} torchaudio=={TORCHAUDIO_VERSION}"
 
-    install_pytorch = f"python -m pip install --upgrade torch=={TORCH_VERSION} torchvision=={TORCHVISION_VERSION} torchaudio=={TORCHAUDIO_VERSION} "
+    if "+cu118" in torver:
+        install_cmd = f"{base_cmd} --index-url https://download.pytorch.org/whl/cu118"
+    elif "+cu" in torver:
+        install_cmd = f"{base_cmd} --index-url https://download.pytorch.org/whl/cu121"
+    elif "+rocm" in torver:
+        install_cmd = f"{base_cmd} --index-url https://download.pytorch.org/whl/rocm6.1"
+    elif "+cpu" in torver:
+        install_cmd = f"{base_cmd} --index-url https://download.pytorch.org/whl/cpu"
+    elif "+cxx11" in torver:
+        intel_extension = "intel-extension-for-pytorch==2.1.10+xpu" if is_linux() else "intel-extension-for-pytorch==2.1.10"
+        install_cmd = f"{base_cmd} {intel_extension} --extra-index-url https://pytorch-extension.intel.com/release-whl/stable/xpu/us/"
+    else:
+        install_cmd = base_cmd
 
-    if is_cuda118:
-        install_pytorch += "--index-url https://download.pytorch.org/whl/cu118"
-    elif is_cuda:
-        install_pytorch += "--index-url https://download.pytorch.org/whl/cu121"
-    elif is_rocm:
-        install_pytorch += "--index-url https://download.pytorch.org/whl/rocm6.1"
-    elif is_cpu:
-        install_pytorch += "--index-url https://download.pytorch.org/whl/cpu"
-    elif is_intel:
-        if is_linux():
-            install_pytorch = "python -m pip install --upgrade torch==2.1.0a0 torchvision==0.16.0a0 torchaudio==2.1.0a0 intel-extension-for-pytorch==2.1.10+xpu --extra-index-url https://pytorch-extension.intel.com/release-whl/stable/xpu/us/"
-        else:
-            install_pytorch = "python -m pip install --upgrade torch==2.1.0a0 torchvision==0.16.0a0 torchaudio==2.1.0a0 intel-extension-for-pytorch==2.1.10 --extra-index-url https://pytorch-extension.intel.com/release-whl/stable/xpu/us/"
-
-    run_cmd(f"{install_pytorch}", assert_success=True, environment=True)
+    run_cmd(install_cmd, assert_success=True, environment=True)
 
 
 def is_installed():
@@ -155,6 +149,11 @@ def check_env():
         sys.exit(1)
 
 
+def get_current_commit():
+    result = run_cmd("git rev-parse HEAD", capture_output=True, environment=True)
+    return result.stdout.decode('utf-8').strip()
+
+
 def clear_cache():
     run_cmd("conda clean -a -y", environment=True)
     run_cmd("python -m pip cache purge", environment=True)
@@ -340,27 +339,84 @@ def install_extensions_requirements():
 def update_requirements(initial_installation=False, pull=True):
     # Create .git directory if missing
     if not os.path.exists(os.path.join(script_dir, ".git")):
-        git_creation_cmd = 'git init -b main && git remote add origin https://github.com/oobabooga/text-generation-webui && git fetch && git symbolic-ref refs/remotes/origin/HEAD refs/remotes/origin/main && git reset --hard origin/main && git branch --set-upstream-to=origin/main'
-        run_cmd(git_creation_cmd, environment=True, assert_success=True)
+        run_cmd(
+            "git init -b main && git remote add origin https://github.com/oobabooga/text-generation-webui && "
+            "git fetch && git symbolic-ref refs/remotes/origin/HEAD refs/remotes/origin/main && "
+            "git reset --hard origin/main && git branch --set-upstream-to=origin/main",
+            environment=True,
+            assert_success=True
+        )
+
+    torver = torch_version()
+    if "+rocm" in torver:
+        requirements_file = "requirements_amd" + ("_noavx2" if not cpu_has_avx2() else "") + ".txt"
+    elif "+cpu" in torver or "+cxx11" in torver:
+        requirements_file = "requirements_cpu_only" + ("_noavx2" if not cpu_has_avx2() else "") + ".txt"
+    elif is_macos():
+        requirements_file = "requirements_apple_" + ("intel" if is_x86_64() else "silicon") + ".txt"
+    else:
+        requirements_file = "requirements" + ("_noavx2" if not cpu_has_avx2() else "") + ".txt"
+
+    # Load state from JSON file
+    state_file = '.installer_state.json'
+    current_commit = get_current_commit()
+    wheels_changed = False
+    if os.path.exists(state_file):
+        with open(state_file, 'r') as f:
+            last_state = json.load(f)
+
+        if 'wheels_changed' in last_state or last_state.get('last_installed_commit') != current_commit:
+            wheels_changed = True
+    else:
+        wheels_changed = True
 
     if pull:
-        print_big_message("Updating the local copy of the repository with \"git pull\"")
+        # Read .whl lines before pulling
+        before_pull_whl_lines = []
+        if os.path.exists(requirements_file):
+            with open(requirements_file, 'r') as f:
+                before_pull_whl_lines = [line for line in f if '.whl' in line]
 
+        print_big_message('Updating the local copy of the repository with "git pull"')
+
+        # Hash files before pulling
         files_to_check = [
             'start_linux.sh', 'start_macos.sh', 'start_windows.bat', 'start_wsl.bat',
             'update_wizard_linux.sh', 'update_wizard_macos.sh', 'update_wizard_windows.bat', 'update_wizard_wsl.bat',
             'one_click.py'
         ]
+        before_hashes = {file: calculate_file_hash(file) for file in files_to_check}
 
-        before_pull_hashes = {file_name: calculate_file_hash(file_name) for file_name in files_to_check}
+        # Perform the git pull
         run_cmd("git pull --autostash", assert_success=True, environment=True)
-        after_pull_hashes = {file_name: calculate_file_hash(file_name) for file_name in files_to_check}
 
-        # Check for differences in installation file hashes
-        for file_name in files_to_check:
-            if before_pull_hashes[file_name] != after_pull_hashes[file_name]:
-                print_big_message(f"File '{file_name}' was updated during 'git pull'. Please run the script again.")
-                exit(1)
+        # Check hashes after pulling
+        after_hashes = {file: calculate_file_hash(file) for file in files_to_check}
+        if os.path.exists(requirements_file):
+            with open(requirements_file, 'r') as f:
+                after_pull_whl_lines = [line for line in f if '.whl' in line]
+
+        wheels_changed = wheels_changed or (before_pull_whl_lines != after_pull_whl_lines)
+
+        # Check for changes to installer files
+        for file in files_to_check:
+            if before_hashes[file] != after_hashes[file]:
+                print_big_message(f"File '{file}' was updated during 'git pull'. Please run the script again.")
+
+                # Save state before exiting
+                current_state = {}
+                if wheels_changed:
+                    current_state['wheels_changed'] = True
+
+                with open(state_file, 'w') as f:
+                    json.dump(current_state, f)
+
+                sys.exit(1)
+
+    # Save current state
+    current_state = {'last_installed_commit': current_commit}
+    with open(state_file, 'w') as f:
+        json.dump(current_state, f)
 
     if os.environ.get("INSTALL_EXTENSIONS", "").lower() in ("yes", "y", "true", "1", "t", "on"):
         install_extensions_requirements()
@@ -369,38 +425,23 @@ def update_requirements(initial_installation=False, pull=True):
     if not initial_installation:
         update_pytorch()
 
-    # Detect the PyTorch version
-    torver = torch_version()
-    is_cuda = '+cu' in torver
-    is_cuda118 = '+cu118' in torver  # 2.1.0+cu118
-    is_rocm = '+rocm' in torver  # 2.0.1+rocm5.4.2
-    is_intel = '+cxx11' in torver  # 2.0.1a0+cxx11.abi
-    is_cpu = '+cpu' in torver  # 2.0.1+cpu
-
-    if is_rocm:
-        base_requirements = "requirements_amd" + ("_noavx2" if not cpu_has_avx2() else "") + ".txt"
-    elif is_cpu or is_intel:
-        base_requirements = "requirements_cpu_only" + ("_noavx2" if not cpu_has_avx2() else "") + ".txt"
-    elif is_macos():
-        base_requirements = "requirements_apple_" + ("intel" if is_x86_64() else "silicon") + ".txt"
-    else:
-        base_requirements = "requirements" + ("_noavx2" if not cpu_has_avx2() else "") + ".txt"
-
-    requirements_file = base_requirements
-
     print_big_message(f"Installing webui requirements from file: {requirements_file}")
     print(f"TORCH: {torver}\n")
 
     # Prepare the requirements file
     textgen_requirements = open(requirements_file).read().splitlines()
-    if is_cuda118:
+
+    if not initial_installation and not wheels_changed:
+        textgen_requirements = [line for line in textgen_requirements if '.whl' not in line]
+
+    if "+cu118" in torver:
         textgen_requirements = [
             req.replace('+cu121', '+cu118').replace('+cu122', '+cu118')
             for req in textgen_requirements
             if "autoawq" not in req.lower()
         ]
 
-    if is_windows() and is_cuda118:  # No flash-attention on Windows for CUDA 11
+    if is_windows() and "+cu118" in torver:  # No flash-attention on Windows for CUDA 11
         textgen_requirements = [req for req in textgen_requirements if 'oobabooga/flash-attention' not in req]
 
     with open('temp_requirements.txt', 'w') as file:
@@ -416,16 +457,9 @@ def update_requirements(initial_installation=False, pull=True):
 
     # Install/update the project requirements
     run_cmd("python -m pip install -r temp_requirements.txt --upgrade", assert_success=True, environment=True)
+
+    # Clean up
     os.remove('temp_requirements.txt')
-
-    # Check for '+cu' or '+rocm' in version string to determine if torch uses CUDA or ROCm. Check for pytorch-cuda as well for backwards compatibility
-    if not any((is_cuda, is_rocm)) and run_cmd("conda list -f pytorch-cuda | grep pytorch-cuda", environment=True, capture_output=True).returncode == 1:
-        clear_cache()
-        return
-
-    if not os.path.exists("repositories/"):
-        os.mkdir("repositories")
-
     clear_cache()
 
 
diff --git a/requirements.txt b/requirements.txt
index 18872431..4ff7a6df 100644
--- a/requirements.txt
+++ b/requirements.txt
@@ -1,4 +1,4 @@
-accelerate==1.2.*
+accelerate==1.3.*
 bitsandbytes==0.45.*
 colorama
 datasets
@@ -32,22 +32,22 @@ sse-starlette==1.6.5
 tiktoken
 
 # llama-cpp-python (CPU only, AVX2)
-https://github.com/oobabooga/llama-cpp-python-cuBLAS-wheels/releases/download/cpu/llama_cpp_python-0.3.6+cpuavx2-cp311-cp311-linux_x86_64.whl; platform_system == "Linux" and platform_machine == "x86_64" and python_version == "3.11"
-https://github.com/oobabooga/llama-cpp-python-cuBLAS-wheels/releases/download/cpu/llama_cpp_python-0.3.6+cpuavx2-cp310-cp310-linux_x86_64.whl; platform_system == "Linux" and platform_machine == "x86_64" and python_version == "3.10"
-https://github.com/oobabooga/llama-cpp-python-cuBLAS-wheels/releases/download/cpu/llama_cpp_python-0.3.6+cpuavx2-cp311-cp311-win_amd64.whl; platform_system == "Windows" and python_version == "3.11"
-https://github.com/oobabooga/llama-cpp-python-cuBLAS-wheels/releases/download/cpu/llama_cpp_python-0.3.6+cpuavx2-cp310-cp310-win_amd64.whl; platform_system == "Windows" and python_version == "3.10"
+https://github.com/oobabooga/llama-cpp-python-cuBLAS-wheels/releases/download/cpu/llama_cpp_python-0.3.7+cpuavx2-cp311-cp311-linux_x86_64.whl; platform_system == "Linux" and platform_machine == "x86_64" and python_version == "3.11"
+https://github.com/oobabooga/llama-cpp-python-cuBLAS-wheels/releases/download/cpu/llama_cpp_python-0.3.7+cpuavx2-cp310-cp310-linux_x86_64.whl; platform_system == "Linux" and platform_machine == "x86_64" and python_version == "3.10"
+https://github.com/oobabooga/llama-cpp-python-cuBLAS-wheels/releases/download/cpu/llama_cpp_python-0.3.7+cpuavx2-cp311-cp311-win_amd64.whl; platform_system == "Windows" and python_version == "3.11"
+https://github.com/oobabooga/llama-cpp-python-cuBLAS-wheels/releases/download/cpu/llama_cpp_python-0.3.7+cpuavx2-cp310-cp310-win_amd64.whl; platform_system == "Windows" and python_version == "3.10"
 
 # llama-cpp-python (CUDA, with GGML_CUDA_FORCE_MMQ)
-https://github.com/oobabooga/llama-cpp-python-cuBLAS-wheels/releases/download/textgen-webui/llama_cpp_python_cuda-0.3.6+cu121-cp311-cp311-win_amd64.whl; platform_system == "Windows" and python_version == "3.11"
-https://github.com/oobabooga/llama-cpp-python-cuBLAS-wheels/releases/download/textgen-webui/llama_cpp_python_cuda-0.3.6+cu121-cp310-cp310-win_amd64.whl; platform_system == "Windows" and python_version == "3.10"
-https://github.com/oobabooga/llama-cpp-python-cuBLAS-wheels/releases/download/textgen-webui/llama_cpp_python_cuda-0.3.6+cu121-cp311-cp311-linux_x86_64.whl; platform_system == "Linux" and platform_machine == "x86_64" and python_version == "3.11"
-https://github.com/oobabooga/llama-cpp-python-cuBLAS-wheels/releases/download/textgen-webui/llama_cpp_python_cuda-0.3.6+cu121-cp310-cp310-linux_x86_64.whl; platform_system == "Linux" and platform_machine == "x86_64" and python_version == "3.10"
+https://github.com/oobabooga/llama-cpp-python-cuBLAS-wheels/releases/download/textgen-webui/llama_cpp_python_cuda-0.3.7+cu121-cp311-cp311-win_amd64.whl; platform_system == "Windows" and python_version == "3.11"
+https://github.com/oobabooga/llama-cpp-python-cuBLAS-wheels/releases/download/textgen-webui/llama_cpp_python_cuda-0.3.7+cu121-cp310-cp310-win_amd64.whl; platform_system == "Windows" and python_version == "3.10"
+https://github.com/oobabooga/llama-cpp-python-cuBLAS-wheels/releases/download/textgen-webui/llama_cpp_python_cuda-0.3.7+cu121-cp311-cp311-linux_x86_64.whl; platform_system == "Linux" and platform_machine == "x86_64" and python_version == "3.11"
+https://github.com/oobabooga/llama-cpp-python-cuBLAS-wheels/releases/download/textgen-webui/llama_cpp_python_cuda-0.3.7+cu121-cp310-cp310-linux_x86_64.whl; platform_system == "Linux" and platform_machine == "x86_64" and python_version == "3.10"
 
 # llama-cpp-python (CUDA, without GGML_CUDA_FORCE_MMQ)
-https://github.com/oobabooga/llama-cpp-python-cuBLAS-wheels/releases/download/textgen-webui/llama_cpp_python_cuda_tensorcores-0.3.6+cu121-cp311-cp311-win_amd64.whl; platform_system == "Windows" and python_version == "3.11"
-https://github.com/oobabooga/llama-cpp-python-cuBLAS-wheels/releases/download/textgen-webui/llama_cpp_python_cuda_tensorcores-0.3.6+cu121-cp310-cp310-win_amd64.whl; platform_system == "Windows" and python_version == "3.10"
-https://github.com/oobabooga/llama-cpp-python-cuBLAS-wheels/releases/download/textgen-webui/llama_cpp_python_cuda_tensorcores-0.3.6+cu121-cp311-cp311-linux_x86_64.whl; platform_system == "Linux" and platform_machine == "x86_64" and python_version == "3.11"
-https://github.com/oobabooga/llama-cpp-python-cuBLAS-wheels/releases/download/textgen-webui/llama_cpp_python_cuda_tensorcores-0.3.6+cu121-cp310-cp310-linux_x86_64.whl; platform_system == "Linux" and platform_machine == "x86_64" and python_version == "3.10"
+https://github.com/oobabooga/llama-cpp-python-cuBLAS-wheels/releases/download/textgen-webui/llama_cpp_python_cuda_tensorcores-0.3.7+cu121-cp311-cp311-win_amd64.whl; platform_system == "Windows" and python_version == "3.11"
+https://github.com/oobabooga/llama-cpp-python-cuBLAS-wheels/releases/download/textgen-webui/llama_cpp_python_cuda_tensorcores-0.3.7+cu121-cp310-cp310-win_amd64.whl; platform_system == "Windows" and python_version == "3.10"
+https://github.com/oobabooga/llama-cpp-python-cuBLAS-wheels/releases/download/textgen-webui/llama_cpp_python_cuda_tensorcores-0.3.7+cu121-cp311-cp311-linux_x86_64.whl; platform_system == "Linux" and platform_machine == "x86_64" and python_version == "3.11"
+https://github.com/oobabooga/llama-cpp-python-cuBLAS-wheels/releases/download/textgen-webui/llama_cpp_python_cuda_tensorcores-0.3.7+cu121-cp310-cp310-linux_x86_64.whl; platform_system == "Linux" and platform_machine == "x86_64" and python_version == "3.10"
 
 # CUDA wheels
 https://github.com/oobabooga/exllamav2/releases/download/v0.2.7/exllamav2-0.2.7+cu121.torch2.4.1-cp311-cp311-win_amd64.whl; platform_system == "Windows" and python_version == "3.11"
diff --git a/requirements_amd.txt b/requirements_amd.txt
index 87ee93d1..e30f30ee 100644
--- a/requirements_amd.txt
+++ b/requirements_amd.txt
@@ -1,4 +1,4 @@
-accelerate==1.2.*
+accelerate==1.3.*
 colorama
 datasets
 einops
@@ -31,14 +31,14 @@ sse-starlette==1.6.5
 tiktoken
 
 # llama-cpp-python (CPU only, AVX2)
-https://github.com/oobabooga/llama-cpp-python-cuBLAS-wheels/releases/download/cpu/llama_cpp_python-0.3.6+cpuavx2-cp311-cp311-linux_x86_64.whl; platform_system == "Linux" and platform_machine == "x86_64" and python_version == "3.11"
-https://github.com/oobabooga/llama-cpp-python-cuBLAS-wheels/releases/download/cpu/llama_cpp_python-0.3.6+cpuavx2-cp310-cp310-linux_x86_64.whl; platform_system == "Linux" and platform_machine == "x86_64" and python_version == "3.10"
-https://github.com/oobabooga/llama-cpp-python-cuBLAS-wheels/releases/download/cpu/llama_cpp_python-0.3.6+cpuavx2-cp311-cp311-win_amd64.whl; platform_system == "Windows" and python_version == "3.11"
-https://github.com/oobabooga/llama-cpp-python-cuBLAS-wheels/releases/download/cpu/llama_cpp_python-0.3.6+cpuavx2-cp310-cp310-win_amd64.whl; platform_system == "Windows" and python_version == "3.10"
+https://github.com/oobabooga/llama-cpp-python-cuBLAS-wheels/releases/download/cpu/llama_cpp_python-0.3.7+cpuavx2-cp311-cp311-linux_x86_64.whl; platform_system == "Linux" and platform_machine == "x86_64" and python_version == "3.11"
+https://github.com/oobabooga/llama-cpp-python-cuBLAS-wheels/releases/download/cpu/llama_cpp_python-0.3.7+cpuavx2-cp310-cp310-linux_x86_64.whl; platform_system == "Linux" and platform_machine == "x86_64" and python_version == "3.10"
+https://github.com/oobabooga/llama-cpp-python-cuBLAS-wheels/releases/download/cpu/llama_cpp_python-0.3.7+cpuavx2-cp311-cp311-win_amd64.whl; platform_system == "Windows" and python_version == "3.11"
+https://github.com/oobabooga/llama-cpp-python-cuBLAS-wheels/releases/download/cpu/llama_cpp_python-0.3.7+cpuavx2-cp310-cp310-win_amd64.whl; platform_system == "Windows" and python_version == "3.10"
 
 # AMD wheels
-https://github.com/oobabooga/llama-cpp-python-cuBLAS-wheels/releases/download/rocm/llama_cpp_python_cuda-0.3.6+rocm6.1.2-cp311-cp311-linux_x86_64.whl; platform_system == "Linux" and platform_machine == "x86_64" and python_version == "3.11"
-https://github.com/oobabooga/llama-cpp-python-cuBLAS-wheels/releases/download/rocm/llama_cpp_python_cuda-0.3.6+rocm6.1.2-cp310-cp310-linux_x86_64.whl; platform_system == "Linux" and platform_machine == "x86_64" and python_version == "3.10"
+https://github.com/oobabooga/llama-cpp-python-cuBLAS-wheels/releases/download/rocm/llama_cpp_python_cuda-0.3.7+rocm6.1.2-cp311-cp311-linux_x86_64.whl; platform_system == "Linux" and platform_machine == "x86_64" and python_version == "3.11"
+https://github.com/oobabooga/llama-cpp-python-cuBLAS-wheels/releases/download/rocm/llama_cpp_python_cuda-0.3.7+rocm6.1.2-cp310-cp310-linux_x86_64.whl; platform_system == "Linux" and platform_machine == "x86_64" and python_version == "3.10"
 https://github.com/oobabooga/exllamav2/releases/download/v0.2.7/exllamav2-0.2.7+rocm6.1.torch2.4.1-cp311-cp311-linux_x86_64.whl; platform_system == "Linux" and platform_machine == "x86_64" and python_version == "3.11"
 https://github.com/oobabooga/exllamav2/releases/download/v0.2.7/exllamav2-0.2.7+rocm6.1.torch2.4.1-cp310-cp310-linux_x86_64.whl; platform_system == "Linux" and platform_machine == "x86_64" and python_version == "3.10"
 https://github.com/oobabooga/exllamav2/releases/download/v0.2.7/exllamav2-0.2.7-py3-none-any.whl; platform_system != "Darwin" and platform_machine != "x86_64"
diff --git a/requirements_amd_noavx2.txt b/requirements_amd_noavx2.txt
index fa2f5ca7..15d25caa 100644
--- a/requirements_amd_noavx2.txt
+++ b/requirements_amd_noavx2.txt
@@ -1,4 +1,4 @@
-accelerate==1.2.*
+accelerate==1.3.*
 colorama
 datasets
 einops
@@ -31,10 +31,10 @@ sse-starlette==1.6.5
 tiktoken
 
 # llama-cpp-python (CPU only, no AVX2)
-https://github.com/oobabooga/llama-cpp-python-cuBLAS-wheels/releases/download/cpu/llama_cpp_python-0.3.6+cpuavx-cp311-cp311-linux_x86_64.whl; platform_system == "Linux" and platform_machine == "x86_64" and python_version == "3.11"
-https://github.com/oobabooga/llama-cpp-python-cuBLAS-wheels/releases/download/cpu/llama_cpp_python-0.3.6+cpuavx-cp310-cp310-linux_x86_64.whl; platform_system == "Linux" and platform_machine == "x86_64" and python_version == "3.10"
-https://github.com/oobabooga/llama-cpp-python-cuBLAS-wheels/releases/download/cpu/llama_cpp_python-0.3.6+cpuavx-cp311-cp311-win_amd64.whl; platform_system == "Windows" and python_version == "3.11"
-https://github.com/oobabooga/llama-cpp-python-cuBLAS-wheels/releases/download/cpu/llama_cpp_python-0.3.6+cpuavx-cp310-cp310-win_amd64.whl; platform_system == "Windows" and python_version == "3.10"
+https://github.com/oobabooga/llama-cpp-python-cuBLAS-wheels/releases/download/cpu/llama_cpp_python-0.3.7+cpuavx-cp311-cp311-linux_x86_64.whl; platform_system == "Linux" and platform_machine == "x86_64" and python_version == "3.11"
+https://github.com/oobabooga/llama-cpp-python-cuBLAS-wheels/releases/download/cpu/llama_cpp_python-0.3.7+cpuavx-cp310-cp310-linux_x86_64.whl; platform_system == "Linux" and platform_machine == "x86_64" and python_version == "3.10"
+https://github.com/oobabooga/llama-cpp-python-cuBLAS-wheels/releases/download/cpu/llama_cpp_python-0.3.7+cpuavx-cp311-cp311-win_amd64.whl; platform_system == "Windows" and python_version == "3.11"
+https://github.com/oobabooga/llama-cpp-python-cuBLAS-wheels/releases/download/cpu/llama_cpp_python-0.3.7+cpuavx-cp310-cp310-win_amd64.whl; platform_system == "Windows" and python_version == "3.10"
 
 # AMD wheels
 https://github.com/oobabooga/exllamav2/releases/download/v0.2.7/exllamav2-0.2.7+rocm6.1.torch2.4.1-cp311-cp311-linux_x86_64.whl; platform_system == "Linux" and platform_machine == "x86_64" and python_version == "3.11"
diff --git a/requirements_apple_intel.txt b/requirements_apple_intel.txt
index e9838295..b614acf4 100644
--- a/requirements_apple_intel.txt
+++ b/requirements_apple_intel.txt
@@ -1,4 +1,4 @@
-accelerate==1.2.*
+accelerate==1.3.*
 colorama
 datasets
 einops
@@ -31,8 +31,8 @@ sse-starlette==1.6.5
 tiktoken
 
 # Mac wheels
-https://github.com/oobabooga/llama-cpp-python-cuBLAS-wheels/releases/download/metal/llama_cpp_python-0.3.6-cp311-cp311-macosx_15_0_x86_64.whl; platform_system == "Darwin" and platform_release >= "24.0.0" and platform_release < "25.0.0" and python_version == "3.11"
-https://github.com/oobabooga/llama-cpp-python-cuBLAS-wheels/releases/download/metal/llama_cpp_python-0.3.6-cp310-cp310-macosx_15_0_x86_64.whl; platform_system == "Darwin" and platform_release >= "24.0.0" and platform_release < "25.0.0" and python_version == "3.10"
-https://github.com/oobabooga/llama-cpp-python-cuBLAS-wheels/releases/download/metal/llama_cpp_python-0.3.6-cp311-cp311-macosx_14_0_x86_64.whl; platform_system == "Darwin" and platform_release >= "23.0.0" and platform_release < "24.0.0" and python_version == "3.11"
-https://github.com/oobabooga/llama-cpp-python-cuBLAS-wheels/releases/download/metal/llama_cpp_python-0.3.6-cp310-cp310-macosx_14_0_x86_64.whl; platform_system == "Darwin" and platform_release >= "23.0.0" and platform_release < "24.0.0" and python_version == "3.10"
+https://github.com/oobabooga/llama-cpp-python-cuBLAS-wheels/releases/download/metal/llama_cpp_python-0.3.7-cp311-cp311-macosx_15_0_x86_64.whl; platform_system == "Darwin" and platform_release >= "24.0.0" and platform_release < "25.0.0" and python_version == "3.11"
+https://github.com/oobabooga/llama-cpp-python-cuBLAS-wheels/releases/download/metal/llama_cpp_python-0.3.7-cp310-cp310-macosx_15_0_x86_64.whl; platform_system == "Darwin" and platform_release >= "24.0.0" and platform_release < "25.0.0" and python_version == "3.10"
+https://github.com/oobabooga/llama-cpp-python-cuBLAS-wheels/releases/download/metal/llama_cpp_python-0.3.7-cp311-cp311-macosx_14_0_x86_64.whl; platform_system == "Darwin" and platform_release >= "23.0.0" and platform_release < "24.0.0" and python_version == "3.11"
+https://github.com/oobabooga/llama-cpp-python-cuBLAS-wheels/releases/download/metal/llama_cpp_python-0.3.7-cp310-cp310-macosx_14_0_x86_64.whl; platform_system == "Darwin" and platform_release >= "23.0.0" and platform_release < "24.0.0" and python_version == "3.10"
 https://github.com/oobabooga/exllamav2/releases/download/v0.2.7/exllamav2-0.2.7-py3-none-any.whl
diff --git a/requirements_apple_silicon.txt b/requirements_apple_silicon.txt
index bef02feb..ca9cc3ac 100644
--- a/requirements_apple_silicon.txt
+++ b/requirements_apple_silicon.txt
@@ -1,4 +1,4 @@
-accelerate==1.2.*
+accelerate==1.3.*
 colorama
 datasets
 einops
@@ -31,10 +31,10 @@ sse-starlette==1.6.5
 tiktoken
 
 # Mac wheels
-https://github.com/oobabooga/llama-cpp-python-cuBLAS-wheels/releases/download/metal/llama_cpp_python-0.3.6-cp311-cp311-macosx_15_0_arm64.whl; platform_system == "Darwin" and platform_release >= "24.0.0" and platform_release < "25.0.0" and python_version == "3.11"
-https://github.com/oobabooga/llama-cpp-python-cuBLAS-wheels/releases/download/metal/llama_cpp_python-0.3.6-cp310-cp310-macosx_15_0_arm64.whl; platform_system == "Darwin" and platform_release >= "24.0.0" and platform_release < "25.0.0" and python_version == "3.10"
-https://github.com/oobabooga/llama-cpp-python-cuBLAS-wheels/releases/download/metal/llama_cpp_python-0.3.6-cp311-cp311-macosx_14_0_arm64.whl; platform_system == "Darwin" and platform_release >= "23.0.0" and platform_release < "24.0.0" and python_version == "3.11"
-https://github.com/oobabooga/llama-cpp-python-cuBLAS-wheels/releases/download/metal/llama_cpp_python-0.3.6-cp310-cp310-macosx_14_0_arm64.whl; platform_system == "Darwin" and platform_release >= "23.0.0" and platform_release < "24.0.0" and python_version == "3.10"
-https://github.com/oobabooga/llama-cpp-python-cuBLAS-wheels/releases/download/metal/llama_cpp_python-0.3.6-cp311-cp311-macosx_13_0_arm64.whl; platform_system == "Darwin" and platform_release >= "22.0.0" and platform_release < "23.0.0" and python_version == "3.11"
-https://github.com/oobabooga/llama-cpp-python-cuBLAS-wheels/releases/download/metal/llama_cpp_python-0.3.6-cp310-cp310-macosx_13_0_arm64.whl; platform_system == "Darwin" and platform_release >= "22.0.0" and platform_release < "23.0.0" and python_version == "3.10"
+https://github.com/oobabooga/llama-cpp-python-cuBLAS-wheels/releases/download/metal/llama_cpp_python-0.3.7-cp311-cp311-macosx_15_0_arm64.whl; platform_system == "Darwin" and platform_release >= "24.0.0" and platform_release < "25.0.0" and python_version == "3.11"
+https://github.com/oobabooga/llama-cpp-python-cuBLAS-wheels/releases/download/metal/llama_cpp_python-0.3.7-cp310-cp310-macosx_15_0_arm64.whl; platform_system == "Darwin" and platform_release >= "24.0.0" and platform_release < "25.0.0" and python_version == "3.10"
+https://github.com/oobabooga/llama-cpp-python-cuBLAS-wheels/releases/download/metal/llama_cpp_python-0.3.7-cp311-cp311-macosx_14_0_arm64.whl; platform_system == "Darwin" and platform_release >= "23.0.0" and platform_release < "24.0.0" and python_version == "3.11"
+https://github.com/oobabooga/llama-cpp-python-cuBLAS-wheels/releases/download/metal/llama_cpp_python-0.3.7-cp310-cp310-macosx_14_0_arm64.whl; platform_system == "Darwin" and platform_release >= "23.0.0" and platform_release < "24.0.0" and python_version == "3.10"
+https://github.com/oobabooga/llama-cpp-python-cuBLAS-wheels/releases/download/metal/llama_cpp_python-0.3.7-cp311-cp311-macosx_13_0_arm64.whl; platform_system == "Darwin" and platform_release >= "22.0.0" and platform_release < "23.0.0" and python_version == "3.11"
+https://github.com/oobabooga/llama-cpp-python-cuBLAS-wheels/releases/download/metal/llama_cpp_python-0.3.7-cp310-cp310-macosx_13_0_arm64.whl; platform_system == "Darwin" and platform_release >= "22.0.0" and platform_release < "23.0.0" and python_version == "3.10"
 https://github.com/oobabooga/exllamav2/releases/download/v0.2.7/exllamav2-0.2.7-py3-none-any.whl
diff --git a/requirements_cpu_only.txt b/requirements_cpu_only.txt
index 32f1a50a..e9a97905 100644
--- a/requirements_cpu_only.txt
+++ b/requirements_cpu_only.txt
@@ -1,4 +1,4 @@
-accelerate==1.2.*
+accelerate==1.3.*
 colorama
 datasets
 einops
@@ -31,7 +31,7 @@ sse-starlette==1.6.5
 tiktoken
 
 # llama-cpp-python (CPU only, AVX2)
-https://github.com/oobabooga/llama-cpp-python-cuBLAS-wheels/releases/download/cpu/llama_cpp_python-0.3.6+cpuavx2-cp311-cp311-linux_x86_64.whl; platform_system == "Linux" and platform_machine == "x86_64" and python_version == "3.11"
-https://github.com/oobabooga/llama-cpp-python-cuBLAS-wheels/releases/download/cpu/llama_cpp_python-0.3.6+cpuavx2-cp310-cp310-linux_x86_64.whl; platform_system == "Linux" and platform_machine == "x86_64" and python_version == "3.10"
-https://github.com/oobabooga/llama-cpp-python-cuBLAS-wheels/releases/download/cpu/llama_cpp_python-0.3.6+cpuavx2-cp311-cp311-win_amd64.whl; platform_system == "Windows" and python_version == "3.11"
-https://github.com/oobabooga/llama-cpp-python-cuBLAS-wheels/releases/download/cpu/llama_cpp_python-0.3.6+cpuavx2-cp310-cp310-win_amd64.whl; platform_system == "Windows" and python_version == "3.10"
+https://github.com/oobabooga/llama-cpp-python-cuBLAS-wheels/releases/download/cpu/llama_cpp_python-0.3.7+cpuavx2-cp311-cp311-linux_x86_64.whl; platform_system == "Linux" and platform_machine == "x86_64" and python_version == "3.11"
+https://github.com/oobabooga/llama-cpp-python-cuBLAS-wheels/releases/download/cpu/llama_cpp_python-0.3.7+cpuavx2-cp310-cp310-linux_x86_64.whl; platform_system == "Linux" and platform_machine == "x86_64" and python_version == "3.10"
+https://github.com/oobabooga/llama-cpp-python-cuBLAS-wheels/releases/download/cpu/llama_cpp_python-0.3.7+cpuavx2-cp311-cp311-win_amd64.whl; platform_system == "Windows" and python_version == "3.11"
+https://github.com/oobabooga/llama-cpp-python-cuBLAS-wheels/releases/download/cpu/llama_cpp_python-0.3.7+cpuavx2-cp310-cp310-win_amd64.whl; platform_system == "Windows" and python_version == "3.10"
diff --git a/requirements_cpu_only_noavx2.txt b/requirements_cpu_only_noavx2.txt
index 938848bf..c4357676 100644
--- a/requirements_cpu_only_noavx2.txt
+++ b/requirements_cpu_only_noavx2.txt
@@ -1,4 +1,4 @@
-accelerate==1.2.*
+accelerate==1.3.*
 colorama
 datasets
 einops
@@ -31,7 +31,7 @@ sse-starlette==1.6.5
 tiktoken
 
 # llama-cpp-python (CPU only, no AVX2)
-https://github.com/oobabooga/llama-cpp-python-cuBLAS-wheels/releases/download/cpu/llama_cpp_python-0.3.6+cpuavx-cp311-cp311-linux_x86_64.whl; platform_system == "Linux" and platform_machine == "x86_64" and python_version == "3.11"
-https://github.com/oobabooga/llama-cpp-python-cuBLAS-wheels/releases/download/cpu/llama_cpp_python-0.3.6+cpuavx-cp310-cp310-linux_x86_64.whl; platform_system == "Linux" and platform_machine == "x86_64" and python_version == "3.10"
-https://github.com/oobabooga/llama-cpp-python-cuBLAS-wheels/releases/download/cpu/llama_cpp_python-0.3.6+cpuavx-cp311-cp311-win_amd64.whl; platform_system == "Windows" and python_version == "3.11"
-https://github.com/oobabooga/llama-cpp-python-cuBLAS-wheels/releases/download/cpu/llama_cpp_python-0.3.6+cpuavx-cp310-cp310-win_amd64.whl; platform_system == "Windows" and python_version == "3.10"
+https://github.com/oobabooga/llama-cpp-python-cuBLAS-wheels/releases/download/cpu/llama_cpp_python-0.3.7+cpuavx-cp311-cp311-linux_x86_64.whl; platform_system == "Linux" and platform_machine == "x86_64" and python_version == "3.11"
+https://github.com/oobabooga/llama-cpp-python-cuBLAS-wheels/releases/download/cpu/llama_cpp_python-0.3.7+cpuavx-cp310-cp310-linux_x86_64.whl; platform_system == "Linux" and platform_machine == "x86_64" and python_version == "3.10"
+https://github.com/oobabooga/llama-cpp-python-cuBLAS-wheels/releases/download/cpu/llama_cpp_python-0.3.7+cpuavx-cp311-cp311-win_amd64.whl; platform_system == "Windows" and python_version == "3.11"
+https://github.com/oobabooga/llama-cpp-python-cuBLAS-wheels/releases/download/cpu/llama_cpp_python-0.3.7+cpuavx-cp310-cp310-win_amd64.whl; platform_system == "Windows" and python_version == "3.10"
diff --git a/requirements_noavx2.txt b/requirements_noavx2.txt
index 69e497e0..40cbc7b0 100644
--- a/requirements_noavx2.txt
+++ b/requirements_noavx2.txt
@@ -1,4 +1,4 @@
-accelerate==1.2.*
+accelerate==1.3.*
 bitsandbytes==0.45.*
 colorama
 datasets
@@ -32,22 +32,22 @@ sse-starlette==1.6.5
 tiktoken
 
 # llama-cpp-python (CPU only, no AVX2)
-https://github.com/oobabooga/llama-cpp-python-cuBLAS-wheels/releases/download/cpu/llama_cpp_python-0.3.6+cpuavx-cp311-cp311-linux_x86_64.whl; platform_system == "Linux" and platform_machine == "x86_64" and python_version == "3.11"
-https://github.com/oobabooga/llama-cpp-python-cuBLAS-wheels/releases/download/cpu/llama_cpp_python-0.3.6+cpuavx-cp310-cp310-linux_x86_64.whl; platform_system == "Linux" and platform_machine == "x86_64" and python_version == "3.10"
-https://github.com/oobabooga/llama-cpp-python-cuBLAS-wheels/releases/download/cpu/llama_cpp_python-0.3.6+cpuavx-cp311-cp311-win_amd64.whl; platform_system == "Windows" and python_version == "3.11"
-https://github.com/oobabooga/llama-cpp-python-cuBLAS-wheels/releases/download/cpu/llama_cpp_python-0.3.6+cpuavx-cp310-cp310-win_amd64.whl; platform_system == "Windows" and python_version == "3.10"
+https://github.com/oobabooga/llama-cpp-python-cuBLAS-wheels/releases/download/cpu/llama_cpp_python-0.3.7+cpuavx-cp311-cp311-linux_x86_64.whl; platform_system == "Linux" and platform_machine == "x86_64" and python_version == "3.11"
+https://github.com/oobabooga/llama-cpp-python-cuBLAS-wheels/releases/download/cpu/llama_cpp_python-0.3.7+cpuavx-cp310-cp310-linux_x86_64.whl; platform_system == "Linux" and platform_machine == "x86_64" and python_version == "3.10"
+https://github.com/oobabooga/llama-cpp-python-cuBLAS-wheels/releases/download/cpu/llama_cpp_python-0.3.7+cpuavx-cp311-cp311-win_amd64.whl; platform_system == "Windows" and python_version == "3.11"
+https://github.com/oobabooga/llama-cpp-python-cuBLAS-wheels/releases/download/cpu/llama_cpp_python-0.3.7+cpuavx-cp310-cp310-win_amd64.whl; platform_system == "Windows" and python_version == "3.10"
 
 # llama-cpp-python (CUDA, with GGML_CUDA_FORCE_MMQ)
-https://github.com/oobabooga/llama-cpp-python-cuBLAS-wheels/releases/download/textgen-webui/llama_cpp_python_cuda-0.3.6+cu121avx-cp311-cp311-win_amd64.whl; platform_system == "Windows" and python_version == "3.11"
-https://github.com/oobabooga/llama-cpp-python-cuBLAS-wheels/releases/download/textgen-webui/llama_cpp_python_cuda-0.3.6+cu121avx-cp310-cp310-win_amd64.whl; platform_system == "Windows" and python_version == "3.10"
-https://github.com/oobabooga/llama-cpp-python-cuBLAS-wheels/releases/download/textgen-webui/llama_cpp_python_cuda-0.3.6+cu121avx-cp311-cp311-linux_x86_64.whl; platform_system == "Linux" and platform_machine == "x86_64" and python_version == "3.11"
-https://github.com/oobabooga/llama-cpp-python-cuBLAS-wheels/releases/download/textgen-webui/llama_cpp_python_cuda-0.3.6+cu121avx-cp310-cp310-linux_x86_64.whl; platform_system == "Linux" and platform_machine == "x86_64" and python_version == "3.10"
+https://github.com/oobabooga/llama-cpp-python-cuBLAS-wheels/releases/download/textgen-webui/llama_cpp_python_cuda-0.3.7+cu121avx-cp311-cp311-win_amd64.whl; platform_system == "Windows" and python_version == "3.11"
+https://github.com/oobabooga/llama-cpp-python-cuBLAS-wheels/releases/download/textgen-webui/llama_cpp_python_cuda-0.3.7+cu121avx-cp310-cp310-win_amd64.whl; platform_system == "Windows" and python_version == "3.10"
+https://github.com/oobabooga/llama-cpp-python-cuBLAS-wheels/releases/download/textgen-webui/llama_cpp_python_cuda-0.3.7+cu121avx-cp311-cp311-linux_x86_64.whl; platform_system == "Linux" and platform_machine == "x86_64" and python_version == "3.11"
+https://github.com/oobabooga/llama-cpp-python-cuBLAS-wheels/releases/download/textgen-webui/llama_cpp_python_cuda-0.3.7+cu121avx-cp310-cp310-linux_x86_64.whl; platform_system == "Linux" and platform_machine == "x86_64" and python_version == "3.10"
 
 # llama-cpp-python (CUDA, without GGML_CUDA_FORCE_MMQ)
-https://github.com/oobabooga/llama-cpp-python-cuBLAS-wheels/releases/download/textgen-webui/llama_cpp_python_cuda_tensorcores-0.3.6+cu121avx-cp311-cp311-win_amd64.whl; platform_system == "Windows" and python_version == "3.11"
-https://github.com/oobabooga/llama-cpp-python-cuBLAS-wheels/releases/download/textgen-webui/llama_cpp_python_cuda_tensorcores-0.3.6+cu121avx-cp310-cp310-win_amd64.whl; platform_system == "Windows" and python_version == "3.10"
-https://github.com/oobabooga/llama-cpp-python-cuBLAS-wheels/releases/download/textgen-webui/llama_cpp_python_cuda_tensorcores-0.3.6+cu121avx-cp311-cp311-linux_x86_64.whl; platform_system == "Linux" and platform_machine == "x86_64" and python_version == "3.11"
-https://github.com/oobabooga/llama-cpp-python-cuBLAS-wheels/releases/download/textgen-webui/llama_cpp_python_cuda_tensorcores-0.3.6+cu121avx-cp310-cp310-linux_x86_64.whl; platform_system == "Linux" and platform_machine == "x86_64" and python_version == "3.10"
+https://github.com/oobabooga/llama-cpp-python-cuBLAS-wheels/releases/download/textgen-webui/llama_cpp_python_cuda_tensorcores-0.3.7+cu121avx-cp311-cp311-win_amd64.whl; platform_system == "Windows" and python_version == "3.11"
+https://github.com/oobabooga/llama-cpp-python-cuBLAS-wheels/releases/download/textgen-webui/llama_cpp_python_cuda_tensorcores-0.3.7+cu121avx-cp310-cp310-win_amd64.whl; platform_system == "Windows" and python_version == "3.10"
+https://github.com/oobabooga/llama-cpp-python-cuBLAS-wheels/releases/download/textgen-webui/llama_cpp_python_cuda_tensorcores-0.3.7+cu121avx-cp311-cp311-linux_x86_64.whl; platform_system == "Linux" and platform_machine == "x86_64" and python_version == "3.11"
+https://github.com/oobabooga/llama-cpp-python-cuBLAS-wheels/releases/download/textgen-webui/llama_cpp_python_cuda_tensorcores-0.3.7+cu121avx-cp310-cp310-linux_x86_64.whl; platform_system == "Linux" and platform_machine == "x86_64" and python_version == "3.10"
 
 # CUDA wheels
 https://github.com/oobabooga/exllamav2/releases/download/v0.2.7/exllamav2-0.2.7+cu121.torch2.4.1-cp311-cp311-win_amd64.whl; platform_system == "Windows" and python_version == "3.11"
diff --git a/requirements_nowheels.txt b/requirements_nowheels.txt
index a034ee61..3d6c922f 100644
--- a/requirements_nowheels.txt
+++ b/requirements_nowheels.txt
@@ -1,4 +1,4 @@
-accelerate==1.2.*
+accelerate==1.3.*
 colorama
 datasets
 einops
diff --git a/start_windows.bat b/start_windows.bat
index 960cfdb7..2e42d6fa 100755
--- a/start_windows.bat
+++ b/start_windows.bat
@@ -41,12 +41,12 @@ if "%conda_exists%" == "F" (
 	mkdir "%INSTALL_DIR%"
 	call curl -Lk "%MINICONDA_DOWNLOAD_URL%" > "%INSTALL_DIR%\miniconda_installer.exe" || ( echo. && echo Miniconda failed to download. && goto end )
 
-	:: Try CertUtil first
+	@rem Try CertUtil first
 	for /f %%a in ('CertUtil -hashfile "%INSTALL_DIR%\miniconda_installer.exe" SHA256 ^| find /i /v " " ^| find /i "%MINICONDA_CHECKSUM%"') do (
 		set "output=%%a"
 	)
 
-	:: If CertUtil fails, try PowerShell
+	@rem If CertUtil fails, try PowerShell
 	if not defined output (
 		for /f %%a in ('powershell -Command "if((Get-FileHash \"%INSTALL_DIR%\miniconda_installer.exe\" -Algorithm SHA256).Hash -eq ''%MINICONDA_CHECKSUM%''){echo true}"') do (
 			set "output=%%a"