diff --git a/qwen3/qwen3_jax/chkpt_utils.py b/qwen3/qwen3_jax/chkpt_utils.py
index 3d2e658..7fcaecb 100644
--- a/qwen3/qwen3_jax/chkpt_utils.py
+++ b/qwen3/qwen3_jax/chkpt_utils.py
@@ -207,6 +207,12 @@ def convert_model_or_layer(
     }
     new_params = {k: None for k in layer_params.keys()}
 
+    # some qwen3 checkpoints store both 'embed_tokens' and 'lm_head' even if the embeddings are tied
+    # in this case, we check that the weights are identical and delete 'lm_head'
+    if cfg.tie_embed and 'lm_head.weight' in torch_params:
+        torch.testing.assert_close(torch_params['lm_head.weight'], torch_params['model.embed_tokens.weight'])
+        del torch_params['lm_head.weight']
+
     def convert_weight_thread(tkey, tweight):
         with jax.default_device(device):
             jweight = convert_weight(tkey, _map_weight(tkey, tweight, custom_transform_map=custom_transform_map), cfg)
diff --git a/qwen3/qwen3_jax/model.py b/qwen3/qwen3_jax/model.py
index 742dc44..0aa500a 100644
--- a/qwen3/qwen3_jax/model.py
+++ b/qwen3/qwen3_jax/model.py
@@ -20,7 +20,7 @@
 from pathlib import Path
 import math
 from functools import partial
-from typing import Callable, Any
+from typing import Callable, Any, Optional
 
 import jax
 import jax.numpy as jnp
@@ -134,6 +134,7 @@ class Config:
     head_dim: int
     vocab_size: int
     max_seq_len: int
+    tie_embed: bool
     # Attention
     causal: bool
     # MoE
@@ -176,6 +177,7 @@ def hf_to_jax_config(hf_config: Any | dict[str, Any]) -> "Config":
         num_layers=_get(hf_config, "num_hidden_layers"),
         head_dim=_get(hf_config, "head_dim"),
         vocab_size=_get(hf_config, "vocab_size"),
+        tie_embed=_get(hf_config, "tie_word_embeddings"),
         norm_eps=_get(hf_config, "rms_norm_eps"),
         moe_experts_per_tok=_get(hf_config, "num_experts_per_tok"),
         moe_num_experts=_get(hf_config, "num_experts"),
@@ -497,7 +499,7 @@ class Weights(_Init):
     layers: list[Layer]
     embedding: jax.Array | ArrayInfo
     gamma_final: jax.Array | ArrayInfo
-    lm_head: jax.Array | ArrayInfo
+    lm_head: Optional[jax.Array | ArrayInfo]
 
     @classmethod
     def abstract(cls, cfg: Config):
@@ -507,7 +509,7 @@ def abstract(cls, cfg: Config):
             layers=layers,
             embedding=ArrayInfo((cfg.vocab_size, cfg.embed), cfg.dtype, ("vocab_in", "vocab_in"), init(0, 1)),
             gamma_final=ArrayInfo((cfg.embed,), cfg.dtype, ("act_embed",), jax.nn.initializers.constant(1.0)),
-            lm_head=ArrayInfo((cfg.embed, cfg.vocab_size), cfg.dtype, ("vocab_in", "vocab_out"), init(1, 0)),
+            lm_head=None if cfg.tie_embed else ArrayInfo((cfg.embed, cfg.vocab_size), cfg.dtype, ("vocab_in", "vocab_out"), init(1, 0)),
         )
 
 
@@ -1082,7 +1084,8 @@ def forward(x: jax.Array, segment_ids: jax.Array, weights: Weights, cfg: Config,
         all_cache_updates.append(cache_updates)
 
     x = rms_norm(x, weights.gamma_final, cfg.norm_eps)  # Final layer norm.
-    logits = einsum("btd,dv->btv", x, weights.lm_head)  # Project to vocabulary size
+    head_weights = weights.embedding.T if cfg.tie_embed else weights.lm_head
+    logits = einsum("btd,dv->btv", x, head_weights)  # Project to vocabulary size
     if is_type(cache, KVCache):
         cache.k, cache.v = [[z[i] for z in all_cache_updates] for i in range(2)]
         additional_tokens = jnp.max(_length_minus_right_padding(segment_ids))