diff --git a/aiserver.py b/aiserver.py
index 0f3a10c5..0384205a 100644
--- a/aiserver.py
+++ b/aiserver.py
@@ -2489,17 +2489,16 @@ def load_model(use_gpu=True, gpu_layers=None, disk_layers=None, initial_load=Fal
                     if(vars.lazy_load):  # torch_lazy_loader.py and low_cpu_mem_usage can't be used at the same time
                         lowmem = {}
                     if(os.path.isdir(vars.custmodpth)):
-                        try:
-                            tokenizer = AutoTokenizer.from_pretrained(vars.custmodpth, revision=vars.revision, cache_dir="cache")
-                        except Exception as e:
-                            pass
                         try:
                             tokenizer = AutoTokenizer.from_pretrained(vars.custmodpth, revision=vars.revision, cache_dir="cache", use_fast=False)
                         except Exception as e:
                             try:
-                                tokenizer = GPT2TokenizerFast.from_pretrained(vars.custmodpth, revision=vars.revision, cache_dir="cache")
+                                tokenizer = AutoTokenizer.from_pretrained(vars.custmodpth, revision=vars.revision, cache_dir="cache")
                             except Exception as e:
-                                tokenizer = GPT2TokenizerFast.from_pretrained("gpt2", revision=vars.revision, cache_dir="cache")
+                                try:
+                                    tokenizer = GPT2TokenizerFast.from_pretrained(vars.custmodpth, revision=vars.revision, cache_dir="cache")
+                                except Exception as e:
+                                    tokenizer = GPT2TokenizerFast.from_pretrained("gpt2", revision=vars.revision, cache_dir="cache")
                         try:
                             model     = AutoModelForCausalLM.from_pretrained(vars.custmodpth, revision=vars.revision, cache_dir="cache", **lowmem)
                         except Exception as e:
@@ -2507,17 +2506,16 @@ def load_model(use_gpu=True, gpu_layers=None, disk_layers=None, initial_load=Fal
                                 raise RuntimeError("One of your GPUs ran out of memory when KoboldAI tried to load your model.")
                             model     = GPTNeoForCausalLM.from_pretrained(vars.custmodpth, revision=vars.revision, cache_dir="cache", **lowmem)
                     elif(os.path.isdir("models/{}".format(vars.model.replace('/', '_')))):
-                        try:
-                            tokenizer = AutoTokenizer.from_pretrained("models/{}".format(vars.model.replace('/', '_')), revision=vars.revision, cache_dir="cache")
-                        except Exception as e:
-                            pass
                         try:
                             tokenizer = AutoTokenizer.from_pretrained("models/{}".format(vars.model.replace('/', '_')), revision=vars.revision, cache_dir="cache", use_fast=False)
                         except Exception as e:
                             try:
-                                tokenizer = GPT2TokenizerFast.from_pretrained("models/{}".format(vars.model.replace('/', '_')), revision=vars.revision, cache_dir="cache")
+                                tokenizer = AutoTokenizer.from_pretrained("models/{}".format(vars.model.replace('/', '_')), revision=vars.revision, cache_dir="cache")
                             except Exception as e:
-                                tokenizer = GPT2TokenizerFast.from_pretrained("gpt2", revision=vars.revision, cache_dir="cache")
+                                try:
+                                    tokenizer = GPT2TokenizerFast.from_pretrained("models/{}".format(vars.model.replace('/', '_')), revision=vars.revision, cache_dir="cache")
+                                except Exception as e:
+                                    tokenizer = GPT2TokenizerFast.from_pretrained("gpt2", revision=vars.revision, cache_dir="cache")
                         try:
                             model     = AutoModelForCausalLM.from_pretrained("models/{}".format(vars.model.replace('/', '_')), revision=vars.revision, cache_dir="cache", **lowmem)
                         except Exception as e:
@@ -2538,17 +2536,16 @@ def load_model(use_gpu=True, gpu_layers=None, disk_layers=None, initial_load=Fal
                             return old_rebuild_tensor(storage, storage_offset, shape, stride)
                         torch._utils._rebuild_tensor = new_rebuild_tensor
 
-                        try:
-                            tokenizer = AutoTokenizer.from_pretrained(vars.model, revision=vars.revision, cache_dir="cache")
-                        except Exception as e:
-                            pass
                         try:
                             tokenizer = AutoTokenizer.from_pretrained(vars.model, revision=vars.revision, cache_dir="cache", use_fast=False)
                         except Exception as e:
                             try:
-                                tokenizer = GPT2TokenizerFast.from_pretrained(vars.model, revision=vars.revision, cache_dir="cache")
+                                tokenizer = AutoTokenizer.from_pretrained(vars.model, revision=vars.revision, cache_dir="cache")
                             except Exception as e:
-                                tokenizer = GPT2TokenizerFast.from_pretrained("gpt2", revision=vars.revision, cache_dir="cache")
+                                try:
+                                    tokenizer = GPT2TokenizerFast.from_pretrained(vars.model, revision=vars.revision, cache_dir="cache")
+                                except Exception as e:
+                                    tokenizer = GPT2TokenizerFast.from_pretrained("gpt2", revision=vars.revision, cache_dir="cache")
                         try:
                             model     = AutoModelForCausalLM.from_pretrained(vars.model, revision=vars.revision, cache_dir="cache", **lowmem)
                         except Exception as e:
diff --git a/tpu_mtj_backend.py b/tpu_mtj_backend.py
index 79582867..743e5fc0 100644
--- a/tpu_mtj_backend.py
+++ b/tpu_mtj_backend.py
@@ -1350,49 +1350,46 @@ def load_model(path: str, driver_version="tpu_driver0.1_dev20210607", hf_checkpo
     print("\n", flush=True)
     with torch_lazy_loader.use_lazy_torch_load(callback=callback, dematerialized_modules=True):
         if(os.path.isdir(vars.custmodpth)):
-            try:
-                tokenizer = AutoTokenizer.from_pretrained(vars.custmodpth, revision=vars.revision, cache_dir="cache")
-            except Exception as e:
-                pass
             try:
                 tokenizer = AutoTokenizer.from_pretrained(vars.custmodpth, revision=vars.revision, cache_dir="cache", use_fast=False)
             except Exception as e:
                 try:
-                    tokenizer = GPT2TokenizerFast.from_pretrained(vars.custmodpth, revision=vars.revision, cache_dir="cache")
+                    tokenizer = AutoTokenizer.from_pretrained(vars.custmodpth, revision=vars.revision, cache_dir="cache")
                 except Exception as e:
-                    tokenizer = GPT2TokenizerFast.from_pretrained("gpt2", revision=vars.revision, cache_dir="cache")
+                    try:
+                        tokenizer = GPT2TokenizerFast.from_pretrained(vars.custmodpth, revision=vars.revision, cache_dir="cache")
+                    except Exception as e:
+                        tokenizer = GPT2TokenizerFast.from_pretrained("gpt2", revision=vars.revision, cache_dir="cache")
             try:
                 model     = AutoModelForCausalLM.from_pretrained(vars.custmodpth, revision=vars.revision, cache_dir="cache")
             except Exception as e:
                 model     = GPTNeoForCausalLM.from_pretrained(vars.custmodpth, revision=vars.revision, cache_dir="cache")
         elif(os.path.isdir("models/{}".format(vars.model.replace('/', '_')))):
-            try:
-                tokenizer = AutoTokenizer.from_pretrained("models/{}".format(vars.model.replace('/', '_')), revision=vars.revision, cache_dir="cache")
-            except Exception as e:
-                pass
             try:
                 tokenizer = AutoTokenizer.from_pretrained("models/{}".format(vars.model.replace('/', '_')), revision=vars.revision, cache_dir="cache", use_fast=False)
             except Exception as e:
                 try:
-                    tokenizer = GPT2TokenizerFast.from_pretrained("models/{}".format(vars.model.replace('/', '_')), revision=vars.revision, cache_dir="cache")
+                    tokenizer = AutoTokenizer.from_pretrained("models/{}".format(vars.model.replace('/', '_')), revision=vars.revision, cache_dir="cache")
                 except Exception as e:
-                    tokenizer = GPT2TokenizerFast.from_pretrained("gpt2", revision=vars.revision, cache_dir="cache")
+                    try:
+                        tokenizer = GPT2TokenizerFast.from_pretrained("models/{}".format(vars.model.replace('/', '_')), revision=vars.revision, cache_dir="cache")
+                    except Exception as e:
+                        tokenizer = GPT2TokenizerFast.from_pretrained("gpt2", revision=vars.revision, cache_dir="cache")
             try:
                 model     = AutoModelForCausalLM.from_pretrained("models/{}".format(vars.model.replace('/', '_')), revision=vars.revision, cache_dir="cache")
             except Exception as e:
                 model     = GPTNeoForCausalLM.from_pretrained("models/{}".format(vars.model.replace('/', '_')), revision=vars.revision, cache_dir="cache")
         else:
-            try:
-                tokenizer = AutoTokenizer.from_pretrained(vars.model, revision=vars.revision, cache_dir="cache")
-            except Exception as e:
-                pass
             try:
                 tokenizer = AutoTokenizer.from_pretrained(vars.model, revision=vars.revision, cache_dir="cache", use_fast=False)
             except Exception as e:
                 try:
-                    tokenizer = GPT2TokenizerFast.from_pretrained(vars.model, revision=vars.revision, cache_dir="cache")
+                    tokenizer = AutoTokenizer.from_pretrained(vars.model, revision=vars.revision, cache_dir="cache")
                 except Exception as e:
-                    tokenizer = GPT2TokenizerFast.from_pretrained("gpt2", revision=vars.revision, cache_dir="cache")
+                    try:
+                        tokenizer = GPT2TokenizerFast.from_pretrained(vars.model, revision=vars.revision, cache_dir="cache")
+                    except Exception as e:
+                        tokenizer = GPT2TokenizerFast.from_pretrained("gpt2", revision=vars.revision, cache_dir="cache")
             try:
                 model     = AutoModelForCausalLM.from_pretrained(vars.model, revision=vars.revision, cache_dir="cache")
             except Exception as e: