Add files using upload-large-folder tool

README.md

@@ -1,20 +1,18 @@
 ---
-library_name: mlx
-license: apache-2.0
-license_link: https://huggingface.co/Qwen/Qwen3-14B/blob/main/LICENSE
-pipeline_tag: text-generation
-base_model: Qwen/Qwen3-14B
 tags:
+- unsloth
 - mlx
+base_model: unsloth/Qwen3-14B
+license: apache-2.0
+pipeline_tag: text-generation
+library_name: mlx
 ---
 
 # mlx-community/Qwen3-14B-4bit-AWQ
 
 This model [mlx-community/Qwen3-14B-4bit-AWQ](https://huggingface.co/mlx-community/Qwen3-14B-4bit-AWQ) was
-converted to MLX format from [Qwen/Qwen3-14B](https://huggingface.co/Qwen/Qwen3-14B)
-using mlx-lm version **0.25.2**.
-
-AWQ Parameters: --bits 4 --group-size 64 --embed-bits 4 --embed-group-size 32 --num-samples 256 --sequence-length 1024 --n-grid 50
+converted to MLX format from [unsloth/Qwen3-14B](https://huggingface.co/unsloth/Qwen3-14B)
+using mlx-lm version **0.26.3**.
 
 ## Use with mlx
 

chat_template.jinja

@@ -18,16 +18,17 @@
 {%- for forward_message in messages %}
     {%- set index = (messages|length - 1) - loop.index0 %}
     {%- set message = messages[index] %}
+    {%- set current_content = message.content if message.content is not none else '' %}
     {%- set tool_start = '<tool_response>' %}
     {%- set tool_start_length = tool_start|length %}
-    {%- set start_of_message = message.content[:tool_start_length] %}
+    {%- set start_of_message = current_content[:tool_start_length] %}
     {%- set tool_end = '</tool_response>' %}
     {%- set tool_end_length = tool_end|length %}
-    {%- set start_pos = (message.content|length) - tool_end_length %}
+    {%- set start_pos = (current_content|length) - tool_end_length %}
     {%- if start_pos < 0 %}
         {%- set start_pos = 0 %}
     {%- endif %}
-    {%- set end_of_message = message.content[start_pos:] %}
+    {%- set end_of_message = current_content[start_pos:] %}
     {%- if ns.multi_step_tool and message.role == "user" and not(start_of_message == tool_start and end_of_message == tool_end) %}
         {%- set ns.multi_step_tool = false %}
         {%- set ns.last_query_index = index %}

config.json

@@ -4,7 +4,6 @@
     ],
     "attention_bias": false,
     "attention_dropout": 0.0,
-    "bos_token_id": 151643,
     "eos_token_id": 151645,
     "head_dim": 128,
     "hidden_act": "silu",
@@ -17,6 +16,7 @@
     "num_attention_heads": 40,
     "num_hidden_layers": 40,
     "num_key_value_heads": 8,
+    "pad_token_id": 151654,
     "quantization": {
         "group_size": 64,
         "bits": 4,
@@ -1350,13 +1350,1347 @@
             "bits": 4
         }
     },
+    "quantization_config": {
+        "group_size": 64,
+        "bits": 4,
+        "model.embed_tokens": {
+            "group_size": 32,
+            "bits": 4
+        },
+        "model.layers.0.self_attn.q_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.0.self_attn.k_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.0.self_attn.v_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.0.self_attn.o_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.0.self_attn.q_norm": false,
+        "model.layers.0.self_attn.k_norm": false,
+        "model.layers.0.self_attn.rope": false,
+        "model.layers.0.mlp.gate_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.0.mlp.down_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.0.mlp.up_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.0.input_layernorm": false,
+        "model.layers.0.post_attention_layernorm": false,
+        "model.layers.1.self_attn.q_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.1.self_attn.k_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.1.self_attn.v_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.1.self_attn.o_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.1.self_attn.q_norm": false,
+        "model.layers.1.self_attn.k_norm": false,
+        "model.layers.1.self_attn.rope": false,
+        "model.layers.1.mlp.gate_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.1.mlp.down_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.1.mlp.up_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.1.input_layernorm": false,
+        "model.layers.1.post_attention_layernorm": false,
+        "model.layers.2.self_attn.q_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.2.self_attn.k_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.2.self_attn.v_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.2.self_attn.o_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.2.self_attn.q_norm": false,
+        "model.layers.2.self_attn.k_norm": false,
+        "model.layers.2.self_attn.rope": false,
+        "model.layers.2.mlp.gate_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.2.mlp.down_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.2.mlp.up_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.2.input_layernorm": false,
+        "model.layers.2.post_attention_layernorm": false,
+        "model.layers.3.self_attn.q_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.3.self_attn.k_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.3.self_attn.v_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.3.self_attn.o_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.3.self_attn.q_norm": false,
+        "model.layers.3.self_attn.k_norm": false,
+        "model.layers.3.self_attn.rope": false,
+        "model.layers.3.mlp.gate_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.3.mlp.down_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.3.mlp.up_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.3.input_layernorm": false,
+        "model.layers.3.post_attention_layernorm": false,
+        "model.layers.4.self_attn.q_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.4.self_attn.k_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.4.self_attn.v_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.4.self_attn.o_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.4.self_attn.q_norm": false,
+        "model.layers.4.self_attn.k_norm": false,
+        "model.layers.4.self_attn.rope": false,
+        "model.layers.4.mlp.gate_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.4.mlp.down_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.4.mlp.up_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.4.input_layernorm": false,
+        "model.layers.4.post_attention_layernorm": false,
+        "model.layers.5.self_attn.q_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.5.self_attn.k_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.5.self_attn.v_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.5.self_attn.o_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.5.self_attn.q_norm": false,
+        "model.layers.5.self_attn.k_norm": false,
+        "model.layers.5.self_attn.rope": false,
+        "model.layers.5.mlp.gate_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.5.mlp.down_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.5.mlp.up_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.5.input_layernorm": false,
+        "model.layers.5.post_attention_layernorm": false,
+        "model.layers.6.self_attn.q_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.6.self_attn.k_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.6.self_attn.v_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.6.self_attn.o_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.6.self_attn.q_norm": false,
+        "model.layers.6.self_attn.k_norm": false,
+        "model.layers.6.self_attn.rope": false,
+        "model.layers.6.mlp.gate_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.6.mlp.down_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.6.mlp.up_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.6.input_layernorm": false,
+        "model.layers.6.post_attention_layernorm": false,
+        "model.layers.7.self_attn.q_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.7.self_attn.k_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.7.self_attn.v_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.7.self_attn.o_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.7.self_attn.q_norm": false,
+        "model.layers.7.self_attn.k_norm": false,
+        "model.layers.7.self_attn.rope": false,
+        "model.layers.7.mlp.gate_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.7.mlp.down_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.7.mlp.up_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.7.input_layernorm": false,
+        "model.layers.7.post_attention_layernorm": false,
+        "model.layers.8.self_attn.q_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.8.self_attn.k_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.8.self_attn.v_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.8.self_attn.o_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.8.self_attn.q_norm": false,
+        "model.layers.8.self_attn.k_norm": false,
+        "model.layers.8.self_attn.rope": false,
+        "model.layers.8.mlp.gate_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.8.mlp.down_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.8.mlp.up_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.8.input_layernorm": false,
+        "model.layers.8.post_attention_layernorm": false,
+        "model.layers.9.self_attn.q_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.9.self_attn.k_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.9.self_attn.v_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.9.self_attn.o_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.9.self_attn.q_norm": false,
+        "model.layers.9.self_attn.k_norm": false,
+        "model.layers.9.self_attn.rope": false,
+        "model.layers.9.mlp.gate_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.9.mlp.down_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.9.mlp.up_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.9.input_layernorm": false,
+        "model.layers.9.post_attention_layernorm": false,
+        "model.layers.10.self_attn.q_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.10.self_attn.k_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.10.self_attn.v_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.10.self_attn.o_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.10.self_attn.q_norm": false,
+        "model.layers.10.self_attn.k_norm": false,
+        "model.layers.10.self_attn.rope": false,
+        "model.layers.10.mlp.gate_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.10.mlp.down_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.10.mlp.up_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.10.input_layernorm": false,
+        "model.layers.10.post_attention_layernorm": false,
+        "model.layers.11.self_attn.q_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.11.self_attn.k_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.11.self_attn.v_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.11.self_attn.o_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.11.self_attn.q_norm": false,
+        "model.layers.11.self_attn.k_norm": false,
+        "model.layers.11.self_attn.rope": false,
+        "model.layers.11.mlp.gate_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.11.mlp.down_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.11.mlp.up_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.11.input_layernorm": false,
+        "model.layers.11.post_attention_layernorm": false,
+        "model.layers.12.self_attn.q_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.12.self_attn.k_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.12.self_attn.v_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.12.self_attn.o_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.12.self_attn.q_norm": false,
+        "model.layers.12.self_attn.k_norm": false,
+        "model.layers.12.self_attn.rope": false,
+        "model.layers.12.mlp.gate_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.12.mlp.down_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.12.mlp.up_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.12.input_layernorm": false,
+        "model.layers.12.post_attention_layernorm": false,
+        "model.layers.13.self_attn.q_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.13.self_attn.k_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.13.self_attn.v_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.13.self_attn.o_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.13.self_attn.q_norm": false,
+        "model.layers.13.self_attn.k_norm": false,
+        "model.layers.13.self_attn.rope": false,
+        "model.layers.13.mlp.gate_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.13.mlp.down_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.13.mlp.up_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.13.input_layernorm": false,
+        "model.layers.13.post_attention_layernorm": false,
+        "model.layers.14.self_attn.q_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.14.self_attn.k_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.14.self_attn.v_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.14.self_attn.o_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.14.self_attn.q_norm": false,
+        "model.layers.14.self_attn.k_norm": false,
+        "model.layers.14.self_attn.rope": false,
+        "model.layers.14.mlp.gate_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.14.mlp.down_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.14.mlp.up_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.14.input_layernorm": false,
+        "model.layers.14.post_attention_layernorm": false,
+        "model.layers.15.self_attn.q_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.15.self_attn.k_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.15.self_attn.v_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.15.self_attn.o_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.15.self_attn.q_norm": false,
+        "model.layers.15.self_attn.k_norm": false,
+        "model.layers.15.self_attn.rope": false,
+        "model.layers.15.mlp.gate_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.15.mlp.down_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.15.mlp.up_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.15.input_layernorm": false,
+        "model.layers.15.post_attention_layernorm": false,
+        "model.layers.16.self_attn.q_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.16.self_attn.k_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.16.self_attn.v_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.16.self_attn.o_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.16.self_attn.q_norm": false,
+        "model.layers.16.self_attn.k_norm": false,
+        "model.layers.16.self_attn.rope": false,
+        "model.layers.16.mlp.gate_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.16.mlp.down_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.16.mlp.up_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.16.input_layernorm": false,
+        "model.layers.16.post_attention_layernorm": false,
+        "model.layers.17.self_attn.q_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.17.self_attn.k_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.17.self_attn.v_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.17.self_attn.o_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.17.self_attn.q_norm": false,
+        "model.layers.17.self_attn.k_norm": false,
+        "model.layers.17.self_attn.rope": false,
+        "model.layers.17.mlp.gate_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.17.mlp.down_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.17.mlp.up_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.17.input_layernorm": false,
+        "model.layers.17.post_attention_layernorm": false,
+        "model.layers.18.self_attn.q_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.18.self_attn.k_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.18.self_attn.v_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.18.self_attn.o_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.18.self_attn.q_norm": false,
+        "model.layers.18.self_attn.k_norm": false,
+        "model.layers.18.self_attn.rope": false,
+        "model.layers.18.mlp.gate_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.18.mlp.down_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.18.mlp.up_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.18.input_layernorm": false,
+        "model.layers.18.post_attention_layernorm": false,
+        "model.layers.19.self_attn.q_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.19.self_attn.k_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.19.self_attn.v_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.19.self_attn.o_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.19.self_attn.q_norm": false,
+        "model.layers.19.self_attn.k_norm": false,
+        "model.layers.19.self_attn.rope": false,
+        "model.layers.19.mlp.gate_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.19.mlp.down_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.19.mlp.up_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.19.input_layernorm": false,
+        "model.layers.19.post_attention_layernorm": false,
+        "model.layers.20.self_attn.q_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.20.self_attn.k_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.20.self_attn.v_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.20.self_attn.o_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.20.self_attn.q_norm": false,
+        "model.layers.20.self_attn.k_norm": false,
+        "model.layers.20.self_attn.rope": false,
+        "model.layers.20.mlp.gate_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.20.mlp.down_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.20.mlp.up_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.20.input_layernorm": false,
+        "model.layers.20.post_attention_layernorm": false,
+        "model.layers.21.self_attn.q_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.21.self_attn.k_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.21.self_attn.v_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.21.self_attn.o_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.21.self_attn.q_norm": false,
+        "model.layers.21.self_attn.k_norm": false,
+        "model.layers.21.self_attn.rope": false,
+        "model.layers.21.mlp.gate_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.21.mlp.down_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.21.mlp.up_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.21.input_layernorm": false,
+        "model.layers.21.post_attention_layernorm": false,
+        "model.layers.22.self_attn.q_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.22.self_attn.k_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.22.self_attn.v_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.22.self_attn.o_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.22.self_attn.q_norm": false,
+        "model.layers.22.self_attn.k_norm": false,
+        "model.layers.22.self_attn.rope": false,
+        "model.layers.22.mlp.gate_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.22.mlp.down_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.22.mlp.up_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.22.input_layernorm": false,
+        "model.layers.22.post_attention_layernorm": false,
+        "model.layers.23.self_attn.q_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.23.self_attn.k_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.23.self_attn.v_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.23.self_attn.o_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.23.self_attn.q_norm": false,
+        "model.layers.23.self_attn.k_norm": false,
+        "model.layers.23.self_attn.rope": false,
+        "model.layers.23.mlp.gate_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.23.mlp.down_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.23.mlp.up_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.23.input_layernorm": false,
+        "model.layers.23.post_attention_layernorm": false,
+        "model.layers.24.self_attn.q_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.24.self_attn.k_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.24.self_attn.v_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.24.self_attn.o_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.24.self_attn.q_norm": false,
+        "model.layers.24.self_attn.k_norm": false,
+        "model.layers.24.self_attn.rope": false,
+        "model.layers.24.mlp.gate_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.24.mlp.down_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.24.mlp.up_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.24.input_layernorm": false,
+        "model.layers.24.post_attention_layernorm": false,
+        "model.layers.25.self_attn.q_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.25.self_attn.k_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.25.self_attn.v_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.25.self_attn.o_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.25.self_attn.q_norm": false,
+        "model.layers.25.self_attn.k_norm": false,
+        "model.layers.25.self_attn.rope": false,
+        "model.layers.25.mlp.gate_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.25.mlp.down_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.25.mlp.up_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.25.input_layernorm": false,
+        "model.layers.25.post_attention_layernorm": false,
+        "model.layers.26.self_attn.q_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.26.self_attn.k_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.26.self_attn.v_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.26.self_attn.o_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.26.self_attn.q_norm": false,
+        "model.layers.26.self_attn.k_norm": false,
+        "model.layers.26.self_attn.rope": false,
+        "model.layers.26.mlp.gate_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.26.mlp.down_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.26.mlp.up_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.26.input_layernorm": false,
+        "model.layers.26.post_attention_layernorm": false,
+        "model.layers.27.self_attn.q_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.27.self_attn.k_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.27.self_attn.v_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.27.self_attn.o_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.27.self_attn.q_norm": false,
+        "model.layers.27.self_attn.k_norm": false,
+        "model.layers.27.self_attn.rope": false,
+        "model.layers.27.mlp.gate_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.27.mlp.down_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.27.mlp.up_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.27.input_layernorm": false,
+        "model.layers.27.post_attention_layernorm": false,
+        "model.layers.28.self_attn.q_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.28.self_attn.k_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.28.self_attn.v_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.28.self_attn.o_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.28.self_attn.q_norm": false,
+        "model.layers.28.self_attn.k_norm": false,
+        "model.layers.28.self_attn.rope": false,
+        "model.layers.28.mlp.gate_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.28.mlp.down_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.28.mlp.up_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.28.input_layernorm": false,
+        "model.layers.28.post_attention_layernorm": false,
+        "model.layers.29.self_attn.q_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.29.self_attn.k_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.29.self_attn.v_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.29.self_attn.o_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.29.self_attn.q_norm": false,
+        "model.layers.29.self_attn.k_norm": false,
+        "model.layers.29.self_attn.rope": false,
+        "model.layers.29.mlp.gate_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.29.mlp.down_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.29.mlp.up_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.29.input_layernorm": false,
+        "model.layers.29.post_attention_layernorm": false,
+        "model.layers.30.self_attn.q_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.30.self_attn.k_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.30.self_attn.v_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.30.self_attn.o_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.30.self_attn.q_norm": false,
+        "model.layers.30.self_attn.k_norm": false,
+        "model.layers.30.self_attn.rope": false,
+        "model.layers.30.mlp.gate_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.30.mlp.down_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.30.mlp.up_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.30.input_layernorm": false,
+        "model.layers.30.post_attention_layernorm": false,
+        "model.layers.31.self_attn.q_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.31.self_attn.k_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.31.self_attn.v_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.31.self_attn.o_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.31.self_attn.q_norm": false,
+        "model.layers.31.self_attn.k_norm": false,
+        "model.layers.31.self_attn.rope": false,
+        "model.layers.31.mlp.gate_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.31.mlp.down_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.31.mlp.up_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.31.input_layernorm": false,
+        "model.layers.31.post_attention_layernorm": false,
+        "model.layers.32.self_attn.q_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.32.self_attn.k_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.32.self_attn.v_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.32.self_attn.o_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.32.self_attn.q_norm": false,
+        "model.layers.32.self_attn.k_norm": false,
+        "model.layers.32.self_attn.rope": false,
+        "model.layers.32.mlp.gate_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.32.mlp.down_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.32.mlp.up_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.32.input_layernorm": false,
+        "model.layers.32.post_attention_layernorm": false,
+        "model.layers.33.self_attn.q_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.33.self_attn.k_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.33.self_attn.v_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.33.self_attn.o_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.33.self_attn.q_norm": false,
+        "model.layers.33.self_attn.k_norm": false,
+        "model.layers.33.self_attn.rope": false,
+        "model.layers.33.mlp.gate_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.33.mlp.down_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.33.mlp.up_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.33.input_layernorm": false,
+        "model.layers.33.post_attention_layernorm": false,
+        "model.layers.34.self_attn.q_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.34.self_attn.k_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.34.self_attn.v_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.34.self_attn.o_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.34.self_attn.q_norm": false,
+        "model.layers.34.self_attn.k_norm": false,
+        "model.layers.34.self_attn.rope": false,
+        "model.layers.34.mlp.gate_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.34.mlp.down_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.34.mlp.up_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.34.input_layernorm": false,
+        "model.layers.34.post_attention_layernorm": false,
+        "model.layers.35.self_attn.q_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.35.self_attn.k_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.35.self_attn.v_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.35.self_attn.o_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.35.self_attn.q_norm": false,
+        "model.layers.35.self_attn.k_norm": false,
+        "model.layers.35.self_attn.rope": false,
+        "model.layers.35.mlp.gate_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.35.mlp.down_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.35.mlp.up_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.35.input_layernorm": false,
+        "model.layers.35.post_attention_layernorm": false,
+        "model.layers.36.self_attn.q_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.36.self_attn.k_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.36.self_attn.v_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.36.self_attn.o_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.36.self_attn.q_norm": false,
+        "model.layers.36.self_attn.k_norm": false,
+        "model.layers.36.self_attn.rope": false,
+        "model.layers.36.mlp.gate_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.36.mlp.down_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.36.mlp.up_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.36.input_layernorm": false,
+        "model.layers.36.post_attention_layernorm": false,
+        "model.layers.37.self_attn.q_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.37.self_attn.k_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.37.self_attn.v_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.37.self_attn.o_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.37.self_attn.q_norm": false,
+        "model.layers.37.self_attn.k_norm": false,
+        "model.layers.37.self_attn.rope": false,
+        "model.layers.37.mlp.gate_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.37.mlp.down_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.37.mlp.up_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.37.input_layernorm": false,
+        "model.layers.37.post_attention_layernorm": false,
+        "model.layers.38.self_attn.q_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.38.self_attn.k_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.38.self_attn.v_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.38.self_attn.o_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.38.self_attn.q_norm": false,
+        "model.layers.38.self_attn.k_norm": false,
+        "model.layers.38.self_attn.rope": false,
+        "model.layers.38.mlp.gate_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.38.mlp.down_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.38.mlp.up_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.38.input_layernorm": false,
+        "model.layers.38.post_attention_layernorm": false,
+        "model.layers.39.self_attn.q_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.39.self_attn.k_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.39.self_attn.v_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.39.self_attn.o_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.39.self_attn.q_norm": false,
+        "model.layers.39.self_attn.k_norm": false,
+        "model.layers.39.self_attn.rope": false,
+        "model.layers.39.mlp.gate_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.39.mlp.down_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.39.mlp.up_proj": {
+            "group_size": 64,
+            "bits": 4
+        },
+        "model.layers.39.input_layernorm": false,
+        "model.layers.39.post_attention_layernorm": false,
+        "model.norm": false,
+        "lm_head": {
+            "group_size": 32,
+            "bits": 4
+        }
+    },
     "rms_norm_eps": 1e-06,
     "rope_scaling": null,
     "rope_theta": 1000000,
     "sliding_window": null,
     "tie_word_embeddings": false,
     "torch_dtype": "bfloat16",
-    "transformers_version": "4.51.0",
+    "transformers_version": "4.51.3",
+    "unsloth_fixed": true,
     "use_cache": true,
     "use_sliding_window": false,
     "vocab_size": 151936

generation_config.json

@@ -1,13 +1,14 @@
 {
-    "bos_token_id": 151643,
-    "do_sample": true,
-    "eos_token_id": [
-        151645,
-        151643
-    ],
-    "pad_token_id": 151643,
-    "temperature": 0.6,
-    "top_k": 20,
-    "top_p": 0.95,
-    "transformers_version": "4.51.0"
-}
+  "bos_token_id": 151643,
+  "do_sample": true,
+  "eos_token_id": [
+    151645,
+    151643
+  ],
+  "max_length": 40960,
+  "pad_token_id": 151654,
+  "temperature": 0.6,
+  "top_k": 20,
+  "top_p": 0.95,
+  "transformers_version": "4.51.3"
+}

model-00001-of-00002.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a34cf9222fabcd23f5818c54655db29c63bb15405e5ed566737df0ff9cd0c84c
+size 5352865510

model-00002-of-00002.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1f0abd78717e47e007ad6c6dacaa763111f29a97fe04b6ac1189cb9856abe3d2
+size 3052272046

special_tokens_map.json

@@ -22,7 +22,7 @@
     "single_word": false
   },
   "pad_token": {
-    "content": "<|endoftext|>",
+    "content": "<|vision_pad|>",
     "lstrip": false,
     "normalized": false,
     "rstrip": false,

tokenizer.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:aeb13307a71acd8fe81861d94ad54ab689df773318809eed3cbe794b4492dae4
+size 11422654

tokenizer_config.json

@@ -227,13 +227,13 @@
     "<|video_pad|>"
   ],
   "bos_token": null,
-  "chat_template": "{%- if tools %}\n    {{- '<|im_start|>system\\n' }}\n    {%- if messages[0].role == 'system' %}\n        {{- messages[0].content + '\\n\\n' }}\n    {%- endif %}\n    {{- \"# Tools\\n\\nYou may call one or more functions to assist with the user query.\\n\\nYou are provided with function signatures within <tools></tools> XML tags:\\n<tools>\" }}\n    {%- for tool in tools %}\n        {{- \"\\n\" }}\n        {{- tool | tojson }}\n    {%- endfor %}\n    {{- \"\\n</tools>\\n\\nFor each function call, return a json object with function name and arguments within <tool_call></tool_call> XML tags:\\n<tool_call>\\n{\\\"name\\\": <function-name>, \\\"arguments\\\": <args-json-object>}\\n</tool_call><|im_end|>\\n\" }}\n{%- else %}\n    {%- if messages[0].role == 'system' %}\n        {{- '<|im_start|>system\\n' + messages[0].content + '<|im_end|>\\n' }}\n    {%- endif %}\n{%- endif %}\n{%- set ns = namespace(multi_step_tool=true, last_query_index=messages|length - 1) %}\n{%- for message in messages[::-1] %}\n    {%- set index = (messages|length - 1) - loop.index0 %}\n    {%- if ns.multi_step_tool and message.role == \"user\" and message.content is string and not(message.content.startswith('<tool_response>') and message.content.endswith('</tool_response>')) %}\n        {%- set ns.multi_step_tool = false %}\n        {%- set ns.last_query_index = index %}\n    {%- endif %}\n{%- endfor %}\n{%- for message in messages %}\n    {%- if message.content is string %}\n        {%- set content = message.content %}\n    {%- else %}\n        {%- set content = '' %}\n    {%- endif %}\n    {%- if (message.role == \"user\") or (message.role == \"system\" and not loop.first) %}\n        {{- '<|im_start|>' + message.role + '\\n' + content + '<|im_end|>' + '\\n' }}\n    {%- elif message.role == \"assistant\" %}\n        {%- set reasoning_content = '' %}\n        {%- if message.reasoning_content is string %}\n            {%- set reasoning_content = message.reasoning_content %}\n        {%- else %}\n            {%- if '</think>' in content %}\n                {%- set reasoning_content = content.split('</think>')[0].rstrip('\\n').split('<think>')[-1].lstrip('\\n') %}\n                {%- set content = content.split('</think>')[-1].lstrip('\\n') %}\n            {%- endif %}\n        {%- endif %}\n        {%- if loop.index0 > ns.last_query_index %}\n            {%- if loop.last or (not loop.last and reasoning_content) %}\n                {{- '<|im_start|>' + message.role + '\\n<think>\\n' + reasoning_content.strip('\\n') + '\\n</think>\\n\\n' + content.lstrip('\\n') }}\n            {%- else %}\n                {{- '<|im_start|>' + message.role + '\\n' + content }}\n            {%- endif %}\n        {%- else %}\n            {{- '<|im_start|>' + message.role + '\\n' + content }}\n        {%- endif %}\n        {%- if message.tool_calls %}\n            {%- for tool_call in message.tool_calls %}\n                {%- if (loop.first and content) or (not loop.first) %}\n                    {{- '\\n' }}\n                {%- endif %}\n                {%- if tool_call.function %}\n                    {%- set tool_call = tool_call.function %}\n                {%- endif %}\n                {{- '<tool_call>\\n{\"name\": \"' }}\n                {{- tool_call.name }}\n                {{- '\", \"arguments\": ' }}\n                {%- if tool_call.arguments is string %}\n                    {{- tool_call.arguments }}\n                {%- else %}\n                    {{- tool_call.arguments | tojson }}\n                {%- endif %}\n                {{- '}\\n</tool_call>' }}\n            {%- endfor %}\n        {%- endif %}\n        {{- '<|im_end|>\\n' }}\n    {%- elif message.role == \"tool\" %}\n        {%- if loop.first or (messages[loop.index0 - 1].role != \"tool\") %}\n            {{- '<|im_start|>user' }}\n        {%- endif %}\n        {{- '\\n<tool_response>\\n' }}\n        {{- content }}\n        {{- '\\n</tool_response>' }}\n        {%- if loop.last or (messages[loop.index0 + 1].role != \"tool\") %}\n            {{- '<|im_end|>\\n' }}\n        {%- endif %}\n    {%- endif %}\n{%- endfor %}\n{%- if add_generation_prompt %}\n    {{- '<|im_start|>assistant\\n' }}\n    {%- if enable_thinking is defined and enable_thinking is false %}\n        {{- '<think>\\n\\n</think>\\n\\n' }}\n    {%- endif %}\n{%- endif %}",
   "clean_up_tokenization_spaces": false,
   "eos_token": "<|im_end|>",
   "errors": "replace",
   "extra_special_tokens": {},
-  "model_max_length": 131072,
-  "pad_token": "<|endoftext|>",
+  "model_max_length": 40960,
+  "pad_token": "<|vision_pad|>",
+  "padding_side": "left",
   "split_special_tokens": false,
   "tokenizer_class": "Qwen2Tokenizer",
   "unk_token": null