From eb495f171f8df7d55c74721eab548a97377539cf Mon Sep 17 00:00:00 2001
From: Daniel Han <danielhanchen@gmail.com>
Date: Fri, 2 May 2025 03:09:44 -0700
Subject: [PATCH] Qwen 3

---
 unsloth/models/llama4.py |  4 +-
 unsloth/models/qwen3.py  | 99 ++--------------------------------------
 2 files changed, 5 insertions(+), 98 deletions(-)

diff --git a/unsloth/models/llama4.py b/unsloth/models/llama4.py
index 9818b3db04..f429922268 100644
--- a/unsloth/models/llama4.py
+++ b/unsloth/models/llama4.py
@@ -12,5 +12,5 @@
 # See the License for the specific language governing permissions and
 # limitations under the License.
 
-from unsloth_studio.models import patch_llama4
-patch_llama4()
+# from unsloth_studio.models import patch_llama4
+# patch_llama4()
diff --git a/unsloth/models/qwen3.py b/unsloth/models/qwen3.py
index c0ceefd10c..ac3ac108fb 100644
--- a/unsloth/models/qwen3.py
+++ b/unsloth/models/qwen3.py
@@ -15,6 +15,7 @@
 from .llama import *
 import os
 from ._utils import __version__
+from unsloth_zoo.utils import Version, _get_dtype
 from .llama import (
     LlamaRotaryEmbedding,
     LlamaLinearScalingRotaryEmbedding,
@@ -28,7 +29,6 @@ try:
         Qwen3ForCausalLM,
     )
 except:
-    from packaging.version import Version
     transformers_version = Version(transformers_version)
     if not transformers_version >= Version("4.50.3"): #TODO: Update when transformers is updated
         raise ImportError(
@@ -51,7 +51,6 @@ except:
     Qwen3SdpaAttention   = Qwen3Attention
     Qwen3FlashAttention2 = Qwen3Attention
 pass
-from unsloth_zoo.utils import Version, _get_dtype
 
 
 def Qwen3Attention_fast_forward(
@@ -286,8 +285,8 @@ def Qwen3Attention_fast_forward_inference(
     Kn = Kn.view(bsz, 1, n_kv_heads, head_dim)#.transpose(1, 2) # we will transpose after normalisation
     Vn = Vn.view(bsz, 1, n_kv_heads, head_dim).transpose(1, 2)
 
-    Qn = fast_rms_layernorm(self.q_norm, Qn)
-    Kn = fast_rms_layernorm(self.k_norm, Kn)
+    Qn = fast_rms_layernorm_inference(self.q_norm, Qn)
+    Kn = fast_rms_layernorm_inference(self.k_norm, Kn)
 
     Qn = Qn.transpose(1, 2)
     Kn = Kn.transpose(1, 2)
@@ -368,98 +367,6 @@ def Qwen3Attention_fast_forward_inference(
     return A, (Kn, Vn)
 pass
 
-# def Qwen3Model_fast_forward_inference(
-#     self,
-#     input_ids,
-#     past_key_values,
-#     position_ids,
-#     attention_mask = None,
-# ):
-#     input_ids = input_ids[:,:self.max_seq_length]
-#     bsz, q_len = input_ids.shape
-#     hd = self.config.hidden_size
-#     mlp_size = self.config.intermediate_size
-
-#     X = self.model.embed_tokens(input_ids)
-#     X = X.to(_get_dtype(self.config.torch_dtype))
-#     bsz, q_len, hd = X.shape
-#     assert(q_len == 1)
-#     # Get saved buffers to reduce memory movement
-#     residual = torch.empty((bsz, q_len, hd), dtype = torch.float32, device = "cuda:0")
-#     _XX = torch.empty((2, bsz, q_len, hd), dtype = torch.float32, device = "cuda:0")
-#     XX, XX2 = _XX[0], _XX[1]
-#     variance = torch.empty((bsz, q_len, 1), dtype = torch.float32, device = "cuda:0")
-#     temp_mlp = torch.empty((2, bsz, 1, mlp_size), dtype = X.dtype, device = "cuda:0")
-#     temp_gate, temp_up = temp_mlp[0], temp_mlp[1]
-
-#     seq_len = past_key_values[0][0].shape[-2]
-#     if bsz != 1:
-#         attention_mask = _prepare_4d_causal_attention_mask_for_sdpa(
-#             attention_mask,
-#             (bsz, q_len),
-#             X,
-#             seq_len,
-#             sliding_window = getattr(self.config, "sliding_window", None),
-#         )
-#     else:
-#         attention_mask = None
-#     pass
-
-#     next_decoder_cache = []
-
-#     for idx, decoder_layer in enumerate(self.model.layers):
-#         residual.copy_(X) # residual = X
-#         X = fast_rms_layernorm_inference(
-#             decoder_layer.input_layernorm,
-#             X,
-#             XX = XX,
-#             XX2 = XX2,
-#             variance = variance,
-#         )
-#         X, present_key_value = Qwen3Attention_fast_forward_inference(
-#             decoder_layer.self_attn,
-#             hidden_states = X,
-#             past_key_value = past_key_values[idx],
-#             position_ids = position_ids,
-#             attention_mask = attention_mask,
-#             do_prefill = not hasattr(decoder_layer.self_attn, "paged_attention"),
-#         )
-#         X += residual
-
-#         residual.copy_(X) # residual = X
-#         X = fast_rms_layernorm_inference(
-#             decoder_layer.post_attention_layernorm,
-#             X,
-#             XX = XX,
-#             XX2 = XX2,
-#             variance = variance,
-#         )
-#         X = fast_swiglu_inference(
-#             decoder_layer.mlp,
-#             X,
-#             temp_gate = temp_gate,
-#             temp_up = temp_up,
-#         )
-#         X += residual
-
-#         next_decoder_cache.append(present_key_value)
-#     pass
-#     X = fast_rms_layernorm_inference(
-#         self.model.norm,
-#         X,
-#         XX = XX,
-#         XX2 = XX2,
-#         variance = variance,
-#     )
-
-#     return BaseModelOutputWithPast(
-#         last_hidden_state = X,
-#         past_key_values = next_decoder_cache,
-#         hidden_states = [],
-#         attentions = [],
-#     )
-# pass
-
 class FastQwen3Model(FastLlamaModel):
 
     @staticmethod