diff --git a/README.md b/README.md
index d27c468e..dc415875 100644
--- a/README.md
+++ b/README.md
@@ -707,9 +707,12 @@ The majority of scripts is licensed under ASL 2.0 (including codes from Diffuser
 ### Working in progress
 
 - __important__ The dependent libraries are updated. Please see [Upgrade](#upgrade) and update the libraries.
-  - transformers, accelerate and huggingface_hub are updated. 
+  - bitsandbytes, transformers, accelerate and huggingface_hub are updated. 
   - If you encounter any issues, please report them.
 
+- `bitsandbytes` is updated to 0.44.0. Now you can use `AdEMAMix8bit` and `PagedAdEMAMix8bit` in the training script. PR [#1640](https://github.com/kohya-ss/sd-scripts/pull/1640) Thanks to sdbds!
+  - There is no abbreviation, so please specify the full path like `--optimizer_type bitsandbytes.optim.AdEMAMix8bit` (not bnb but bitsandbytes).
+
 - Fixed a bug in the cache of latents. When `flip_aug`, `alpha_mask`, and `random_crop` are different in multiple subsets in the dataset configuration file (.toml), the last subset is used instead of reflecting them correctly.
 
 - Fixed an issue where the timesteps in the batch were the same when using Huber loss. PR [#1628](https://github.com/kohya-ss/sd-scripts/pull/1628) Thanks to recris!
diff --git a/library/train_util.py b/library/train_util.py
index 17dd447e..3768b605 100644
--- a/library/train_util.py
+++ b/library/train_util.py
@@ -3312,7 +3312,11 @@ def add_optimizer_arguments(parser: argparse.ArgumentParser):
         "--optimizer_type",
         type=str,
         default="",
-        help="Optimizer to use / オプティマイザの種類: AdamW (default), AdamW8bit, PagedAdamW, PagedAdamW8bit, PagedAdamW32bit, Lion8bit, PagedLion8bit, Lion, SGDNesterov, SGDNesterov8bit, DAdaptation(DAdaptAdamPreprint), DAdaptAdaGrad, DAdaptAdam, DAdaptAdan, DAdaptAdanIP, DAdaptLion, DAdaptSGD, AdaFactor",
+        help="Optimizer to use / オプティマイザの種類: AdamW (default), AdamW8bit, PagedAdamW, PagedAdamW8bit, PagedAdamW32bit, "
+        "Lion8bit, PagedLion8bit, Lion, SGDNesterov, SGDNesterov8bit, "
+        "DAdaptation(DAdaptAdamPreprint), DAdaptAdaGrad, DAdaptAdam, DAdaptAdan, DAdaptAdanIP, DAdaptLion, DAdaptSGD, "
+        "AdaFactor. "
+        "Also, you can use any optimizer by specifying the full path to the class, like 'bitsandbytes.optim.AdEMAMix8bit' or 'bitsandbytes.optim.PagedAdEMAMix8bit'.",
     )
 
     # backward compatibility
@@ -4364,7 +4368,7 @@ def resume_from_local_or_hf_if_specified(accelerator, args):
 
 
 def get_optimizer(args, trainable_params):
-    # "Optimizer to use: AdamW, AdamW8bit, Lion, SGDNesterov, SGDNesterov8bit, PagedAdamW, PagedAdamW8bit, PagedAdamW32bit, Lion8bit, PagedLion8bit, DAdaptation(DAdaptAdamPreprint), DAdaptAdaGrad, DAdaptAdam, DAdaptAdan, DAdaptAdanIP, DAdaptLion, DAdaptSGD, Adafactor"
+    # "Optimizer to use: AdamW, AdamW8bit, Lion, SGDNesterov, SGDNesterov8bit, PagedAdamW, PagedAdamW8bit, PagedAdamW32bit, Lion8bit, PagedLion8bit, AdEMAMix8bit, PagedAdEMAMix8bit, DAdaptation(DAdaptAdamPreprint), DAdaptAdaGrad, DAdaptAdam, DAdaptAdan, DAdaptAdanIP, DAdaptLion, DAdaptSGD, Adafactor"
 
     optimizer_type = args.optimizer_type
     if args.use_8bit_adam:
@@ -4417,6 +4421,7 @@ def get_optimizer(args, trainable_params):
 
     lr = args.learning_rate
     optimizer = None
+    optimizer_class = None
 
     if optimizer_type == "Lion".lower():
         try:
@@ -4474,7 +4479,8 @@ def get_optimizer(args, trainable_params):
                     "No PagedLion8bit. The version of bitsandbytes installed seems to be old. Please install 0.39.0 or later. / PagedLion8bitが定義されていません。インストールされているbitsandbytesのバージョンが古いようです。0.39.0以上をインストールしてください"
                 )
 
-        optimizer = optimizer_class(trainable_params, lr=lr, **optimizer_kwargs)
+        if optimizer_class is not None:
+            optimizer = optimizer_class(trainable_params, lr=lr, **optimizer_kwargs)
 
     elif optimizer_type == "PagedAdamW".lower():
         logger.info(f"use PagedAdamW optimizer | {optimizer_kwargs}")
@@ -4749,6 +4755,7 @@ def get_optimizer(args, trainable_params):
         logger.info(f"wrap optimizer with ScheduleFreeWrapper | {schedulefree_wrapper_kwargs}")
     """
 
+    # for logging
     optimizer_name = optimizer_class.__module__ + "." + optimizer_class.__name__
     optimizer_args = ",".join([f"{k}={v}" for k, v in optimizer_kwargs.items()])
 
diff --git a/requirements.txt b/requirements.txt
index bab53f20..0dd1c69c 100644
--- a/requirements.txt
+++ b/requirements.txt
@@ -6,7 +6,7 @@ ftfy==6.1.1
 opencv-python==4.8.1.78
 einops==0.7.0
 pytorch-lightning==1.9.0
-bitsandbytes==0.43.3
+bitsandbytes==0.44.0
 prodigyopt==1.0
 lion-pytorch==0.0.6
 schedulefree==1.2.7