Upload NVFP4 quantized Qwen3-VLTO-32B-Instruct model

.gitattributes

@@ -33,3 +33,4 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 *.zip filter=lfs diff=lfs merge=lfs -text
 *.zst filter=lfs diff=lfs merge=lfs -text
 *tfevents* filter=lfs diff=lfs merge=lfs -text
+tokenizer.json filter=lfs diff=lfs merge=lfs -text

README.md

@@ -0,0 +1,190 @@
+---
+language:
+- en
+license: apache-2.0
+tags:
+- text-generation
+- quantization
+- nvfp4
+- nvidia
+- dgx-spark
+- blackwell
+- model_hub_mixin
+- pytorch_model_hub_mixin
+base_model: qingy2024/Qwen3-VLTO-32B-Instruct
+inference: false
+---
+
+# Qwen3-VLTO-32B-Instruct-NVFP4
+
+This is an NVFP4 quantized version of [qingy2024/Qwen3-VLTO-32B-Instruct](https://huggingface.co/qingy2024/Qwen3-VLTO-32B-Instruct), optimized for NVIDIA DGX Spark systems with Blackwell GB10 GPUs.
+
+## Model Description
+
+- **Base Model:** qingy2024/Qwen3-VLTO-32B-Instruct
+- **Quantization Format:** NVFP4 (4-bit floating point)
+- **Target Hardware:** NVIDIA DGX Spark (Grace Blackwell Superchip)
+- **Quantization Tool:** NVIDIA TensorRT Model Optimizer v0.35.1
+- **Model Size:** Approximately 20 GB (68% reduction from BF16)
+
+## Performance Characteristics
+
+### Memory Efficiency
+
+| Model Version | Memory Usage | Reduction |
+|--------------|--------------|-----------|
+| BF16 (Original) | 61.03 GB | Baseline |
+| NVFP4 (This model) | 19.42 GB | 68.2% |
+
+### Inference Speed
+
+| Model Version | Throughput | Relative Performance |
+|--------------|------------|---------------------|
+| BF16 (Original) | 3.65 tokens/s | Baseline |
+| NVFP4 (This model) | 9.99 tokens/s | 2.74x faster |
+
+**Test Configuration:**
+- Hardware: NVIDIA DGX Spark GB10
+- Framework: vLLM 0.10.2
+- Max Model Length: 8192 tokens
+- GPU Memory Utilization: 90%
+
+## Quantization Details
+
+### NVFP4 Format
+
+NVFP4 is NVIDIA's 4-bit floating point quantization format featuring:
+- **Two-level scaling:** E4M3 FP8 scaling per 16-value block + global FP32 tensor scale
+- **Hardware acceleration:** Optimized for Tensor Cores on Blackwell GB10 GPUs
+- **Group size:** 16
+- **Minimal accuracy degradation:** Less than 1% vs original model
+- **Excluded modules:** lm_head (kept in higher precision)
+
+### Calibration
+
+- **Dataset:** C4 (Colossal Clean Crawled Corpus)
+- **Calibration samples:** 512
+- **Maximum sequence length:** 2048 tokens
+- **Method:** Post-training quantization with activation calibration
+
+## Usage
+
+### Requirements
+
+- NVIDIA DGX Spark or compatible Blackwell GPU
+- vLLM >= 0.6.5
+- nvidia-modelopt[hf]
+
+### Loading the Model
+
+**IMPORTANT:** This model must be loaded with vLLM using the `modelopt` quantization parameter. Standard HuggingFace `AutoModelForCausalLM` will not work.
+
+```python
+from vllm import LLM, SamplingParams
+
+# Load NVFP4 quantized model
+llm = LLM(
+    model="Ex0bit/Qwen3-VLTO-32B-Instruct-NVFP4",
+    quantization="modelopt",  # Required for NVFP4
+    trust_remote_code=True,
+    gpu_memory_utilization=0.9
+)
+
+# Generate
+sampling_params = SamplingParams(temperature=0.7, top_p=0.9, max_tokens=256)
+outputs = llm.generate(["Explain quantum computing in simple terms:"], sampling_params)
+print(outputs[0].outputs[0].text)
+```
+
+### Environment Variables
+
+You can optionally set:
+- `HF_CACHE_DIR`: Override HuggingFace cache location
+
+## Limitations
+
+- **Hardware specific:** Optimized for NVIDIA Blackwell architecture (GB10)
+- **vLLM required:** Cannot be loaded with standard transformers library
+- **Quantization artifacts:** Minor precision loss (<1%) compared to BF16 original
+
+## Intended Use
+
+This model is intended for:
+- High-throughput inference on NVIDIA DGX Spark systems
+- Production deployments requiring memory-efficient models
+- Research on quantization techniques for large language models
+
+## Training and Quantization
+
+### Base Model Training
+
+See the [original model card](https://huggingface.co/qingy2024/Qwen3-VLTO-32B-Instruct) for base model training details.
+
+### Quantization Process
+
+1. **Model Loading:** Original model loaded in BF16 precision
+2. **Calibration:** 512 samples from C4 dataset for activation statistics
+3. **Quantization:** NVFP4 format applied using NVIDIA modelopt
+4. **Export:** Saved in HuggingFace safetensors format
+
+**Quantization Time:** Approximately 60-90 minutes on DGX Spark
+
+## Evaluation
+
+### Test Results
+
+All 5 inference tests passed successfully:
+- Technical explanation generation
+- Code generation
+- Mathematical reasoning
+- Creative writing
+- Instruction following
+
+**Average performance:** 9.99 tokens/s on DGX Spark GB10
+
+## Citation
+
+If you use this quantized model, please cite:
+
+```bibtex
+@misc{qwen3vlto32b-nvfp4,
+  author = {Ex0bit},
+  title = {Qwen3-VLTO-32B-Instruct-NVFP4: NVFP4 Quantized Model for DGX Spark},
+  year = {2025},
+  publisher = {HuggingFace},
+  howpublished = {\url{https://huggingface.co/Ex0bit/Qwen3-VLTO-32B-Instruct-NVFP4}},
+}
+```
+
+And the original base model:
+
+```bibtex
+@misc{qingy2024qwen3vlto,
+  author = {qingy2024},
+  title = {Qwen3-VLTO-32B-Instruct},
+  year = {2024},
+  publisher = {HuggingFace},
+  howpublished = {\url{https://huggingface.co/qingy2024/Qwen3-VLTO-32B-Instruct}},
+}
+```
+
+## References
+
+- [NVIDIA TensorRT Model Optimizer](https://github.com/NVIDIA/TensorRT-Model-Optimizer)
+- [vLLM Documentation](https://docs.vllm.ai/)
+- [NVIDIA DGX Spark Documentation](https://docs.nvidia.com/dgx-spark/)
+- [Quantization GitHub Repository](https://github.com/Ex0bit/nvfp4-quantization)
+
+## License
+
+This quantized model inherits the license from the base model. Please refer to the [original model's license](https://huggingface.co/qingy2024/Qwen3-VLTO-32B-Instruct) for details.
+
+## Model Card Authors
+
+- Ex0bit (@Ex0bit)
+
+## Acknowledgments
+
+- NVIDIA for TensorRT Model Optimizer and DGX Spark hardware
+- qingy2024 for the base Qwen3-VLTO-32B-Instruct model
+- The vLLM team for high-performance inference framework

added_tokens.json

@@ -0,0 +1,28 @@
+{
+  "</think>": 151668,
+  "</tool_call>": 151658,
+  "</tool_response>": 151666,
+  "<think>": 151667,
+  "<tool_call>": 151657,
+  "<tool_response>": 151665,
+  "<|box_end|>": 151649,
+  "<|box_start|>": 151648,
+  "<|endoftext|>": 151643,
+  "<|file_sep|>": 151664,
+  "<|fim_middle|>": 151660,
+  "<|fim_pad|>": 151662,
+  "<|fim_prefix|>": 151659,
+  "<|fim_suffix|>": 151661,
+  "<|im_end|>": 151645,
+  "<|im_start|>": 151644,
+  "<|image_pad|>": 151655,
+  "<|object_ref_end|>": 151647,
+  "<|object_ref_start|>": 151646,
+  "<|quad_end|>": 151651,
+  "<|quad_start|>": 151650,
+  "<|repo_name|>": 151663,
+  "<|video_pad|>": 151656,
+  "<|vision_end|>": 151653,
+  "<|vision_pad|>": 151654,
+  "<|vision_start|>": 151652
+}

chat_template.jinja

@@ -0,0 +1,89 @@
+{%- if tools %}
+    {{- '<|im_start|>system\n' }}
+    {%- if messages[0].role == 'system' %}
+        {{- messages[0].content + '\n\n' }}
+    {%- endif %}
+    {{- "# 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>" }}
+    {%- for tool in tools %}
+        {{- "\n" }}
+        {{- tool | tojson }}
+    {%- endfor %}
+    {{- "\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" }}
+{%- else %}
+    {%- if messages[0].role == 'system' %}
+        {{- '<|im_start|>system\n' + messages[0].content + '<|im_end|>\n' }}
+    {%- endif %}
+{%- endif %}
+{%- set ns = namespace(multi_step_tool=true, last_query_index=messages|length - 1) %}
+{%- for message in messages[::-1] %}
+    {%- set index = (messages|length - 1) - loop.index0 %}
+    {%- 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>')) %}
+        {%- set ns.multi_step_tool = false %}
+        {%- set ns.last_query_index = index %}
+    {%- endif %}
+{%- endfor %}
+{%- for message in messages %}
+    {%- if message.content is string %}
+        {%- set content = message.content %}
+    {%- else %}
+        {%- set content = '' %}
+    {%- endif %}
+    {%- if (message.role == "user") or (message.role == "system" and not loop.first) %}
+        {{- '<|im_start|>' + message.role + '\n' + content + '<|im_end|>' + '\n' }}
+    {%- elif message.role == "assistant" %}
+        {%- set reasoning_content = '' %}
+        {%- if message.reasoning_content is string %}
+            {%- set reasoning_content = message.reasoning_content %}
+        {%- else %}
+            {%- if '</think>' in content %}
+                {%- set reasoning_content = content.split('</think>')[0].rstrip('\n').split('<think>')[-1].lstrip('\n') %}
+                {%- set content = content.split('</think>')[-1].lstrip('\n') %}
+            {%- endif %}
+        {%- endif %}
+        {%- if loop.index0 > ns.last_query_index %}
+            {%- if loop.last or (not loop.last and reasoning_content) %}
+                {{- '<|im_start|>' + message.role + '\n<think>\n' + reasoning_content.strip('\n') + '\n</think>\n\n' + content.lstrip('\n') }}
+            {%- else %}
+                {{- '<|im_start|>' + message.role + '\n' + content }}
+            {%- endif %}
+        {%- else %}
+            {{- '<|im_start|>' + message.role + '\n' + content }}
+        {%- endif %}
+        {%- if message.tool_calls %}
+            {%- for tool_call in message.tool_calls %}
+                {%- if (loop.first and content) or (not loop.first) %}
+                    {{- '\n' }}
+                {%- endif %}
+                {%- if tool_call.function %}
+                    {%- set tool_call = tool_call.function %}
+                {%- endif %}
+                {{- '<tool_call>\n{"name": "' }}
+                {{- tool_call.name }}
+                {{- '", "arguments": ' }}
+                {%- if tool_call.arguments is string %}
+                    {{- tool_call.arguments }}
+                {%- else %}
+                    {{- tool_call.arguments | tojson }}
+                {%- endif %}
+                {{- '}\n</tool_call>' }}
+            {%- endfor %}
+        {%- endif %}
+        {{- '<|im_end|>\n' }}
+    {%- elif message.role == "tool" %}
+        {%- if loop.first or (messages[loop.index0 - 1].role != "tool") %}
+            {{- '<|im_start|>user' }}
+        {%- endif %}
+        {{- '\n<tool_response>\n' }}
+        {{- content }}
+        {{- '\n</tool_response>' }}
+        {%- if loop.last or (messages[loop.index0 + 1].role != "tool") %}
+            {{- '<|im_end|>\n' }}
+        {%- endif %}
+    {%- endif %}
+{%- endfor %}
+{%- if add_generation_prompt %}
+    {{- '<|im_start|>assistant\n' }}
+    {%- if enable_thinking is defined and enable_thinking is false %}
+        {{- '<think>\n\n</think>\n\n' }}
+    {%- endif %}
+{%- endif %}

config.json

@@ -0,0 +1,127 @@
+{
+    "architectures": [
+        "Qwen3ForCausalLM"
+    ],
+    "attention_bias": false,
+    "attention_dropout": 0.0,
+    "bos_token_id": 151643,
+    "dtype": "bfloat16",
+    "eos_token_id": 151645,
+    "head_dim": 128,
+    "hidden_act": "silu",
+    "hidden_size": 5120,
+    "initializer_range": 0.02,
+    "intermediate_size": 25600,
+    "layer_types": [
+        "full_attention",
+        "full_attention",
+        "full_attention",
+        "full_attention",
+        "full_attention",
+        "full_attention",
+        "full_attention",
+        "full_attention",
+        "full_attention",
+        "full_attention",
+        "full_attention",
+        "full_attention",
+        "full_attention",
+        "full_attention",
+        "full_attention",
+        "full_attention",
+        "full_attention",
+        "full_attention",
+        "full_attention",
+        "full_attention",
+        "full_attention",
+        "full_attention",
+        "full_attention",
+        "full_attention",
+        "full_attention",
+        "full_attention",
+        "full_attention",
+        "full_attention",
+        "full_attention",
+        "full_attention",
+        "full_attention",
+        "full_attention",
+        "full_attention",
+        "full_attention",
+        "full_attention",
+        "full_attention",
+        "full_attention",
+        "full_attention",
+        "full_attention",
+        "full_attention",
+        "full_attention",
+        "full_attention",
+        "full_attention",
+        "full_attention",
+        "full_attention",
+        "full_attention",
+        "full_attention",
+        "full_attention",
+        "full_attention",
+        "full_attention",
+        "full_attention",
+        "full_attention",
+        "full_attention",
+        "full_attention",
+        "full_attention",
+        "full_attention",
+        "full_attention",
+        "full_attention",
+        "full_attention",
+        "full_attention",
+        "full_attention",
+        "full_attention",
+        "full_attention",
+        "full_attention"
+    ],
+    "max_position_embeddings": 40960,
+    "max_window_layers": 64,
+    "model_type": "qwen3",
+    "num_attention_heads": 64,
+    "num_hidden_layers": 64,
+    "num_key_value_heads": 8,
+    "rms_norm_eps": 1e-06,
+    "rope_scaling": null,
+    "rope_theta": 1000000,
+    "sliding_window": null,
+    "tie_word_embeddings": false,
+    "torch_dtype": "bfloat16",
+    "transformers_version": "4.55.4",
+    "use_cache": true,
+    "use_sliding_window": false,
+    "vocab_size": 151936,
+    "quantization_config": {
+        "config_groups": {
+            "group_0": {
+                "input_activations": {
+                    "dynamic": false,
+                    "num_bits": 4,
+                    "type": "float",
+                    "group_size": 16
+                },
+                "weights": {
+                    "dynamic": false,
+                    "num_bits": 4,
+                    "type": "float",
+                    "group_size": 16
+                },
+                "targets": [
+                    "Linear"
+                ]
+            }
+        },
+        "ignore": [
+            "lm_head"
+        ],
+        "quant_algo": "NVFP4",
+        "producer": {
+            "name": "modelopt",
+            "version": "0.35.1"
+        },
+        "quant_method": "modelopt"
+    }
+}

generation_config.json

@@ -0,0 +1,13 @@
+{
+  "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": "4.55.4"
+}

hf_quant_config.json

@@ -0,0 +1,14 @@
+{
+    "producer": {
+        "name": "modelopt",
+        "version": "0.35.1"
+    },
+    "quantization": {
+        "quant_algo": "NVFP4",
+        "kv_cache_quant_algo": null,
+        "group_size": 16,
+        "exclude_modules": [
+            "lm_head"
+        ]
+    }
+}

model-00001-of-00005.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:90c90e79818e81817d212d02d01d6a2b7a69831d5335904da03f4347cdb30cb0
+size 4974152288

model-00002-of-00005.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6b40ff0dad957d87175cee08c1e797a83ed1ec53f2148fb44754fd3513f9d08e
+size 4937268176

model-00003-of-00005.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3153f2c7a65178315ffaf1fa10e14e40012474103cc5b1f24b907561b414d3d6
+size 4937268176

model-00004-of-00005.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:71d34b4436543bc61f10e24b1562252313d3f56528b8da6d36d86a72672673e1
+size 4261878016

model-00005-of-00005.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7cf1aab5d6a34929aaf97ac0d6028a8af456567ce32f3914783356875457af7a
+size 1555824768

quantization_metadata.json

@@ -0,0 +1,29 @@
+{
+  "model_name": "qingy2024/Qwen3-VLTO-32B-Instruct",
+  "quantization_format": "NVFP4",
+  "calibration_samples": 512,
+  "original_memory_gb": "65.52",
+  "quantized_memory_gb": "65.53",
+  "memory_reduction_percent": "-0.0",
+  "total_parameters_billions": "32.76",
+  "target_device": "DGX Spark (GB10)",
+  "export_path": "/workspace/quantized-output/Qwen3-VLTO-32B-Instruct-NVFP4",
+  "quantization_config": {
+    "model_name": "qingy2024/Qwen3-VLTO-32B-Instruct",
+    "quantization_format": "NVFP4",
+    "calibration_samples": 512,
+    "calibration_max_length": 2048,
+    "calibration_min_length": 100,
+    "calibration_dataset": "allenai/c4",
+    "calibration_dataset_config": "en",
+    "calibration_dataset_split": "train",
+    "target_device": "DGX Spark (GB10)",
+    "hf_cache_dir": "~/.cache/huggingface",
+    "output_dir": "./quantized-output/Qwen3-VLTO-32B-Instruct-NVFP4",
+    "log_dir": "./logs",
+    "torch_dtype": "bfloat16",
+    "device_map": "auto",
+    "trust_remote_code": true
+  },
+  "created_at": "2025-11-03T03:33:54.755646"
+}

special_tokens_map.json

@@ -0,0 +1,31 @@
+{
+  "additional_special_tokens": [
+    "<|im_start|>",
+    "<|im_end|>",
+    "<|object_ref_start|>",
+    "<|object_ref_end|>",
+    "<|box_start|>",
+    "<|box_end|>",
+    "<|quad_start|>",
+    "<|quad_end|>",
+    "<|vision_start|>",
+    "<|vision_end|>",
+    "<|vision_pad|>",
+    "<|image_pad|>",
+    "<|video_pad|>"
+  ],
+  "eos_token": {
+    "content": "<|im_end|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": {
+    "content": "<|endoftext|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  }
+}

tokenizer.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c0acdaba32b920d640afb36af4396c91974e074735636e4016d17a8ed9c03730
+size 11422753

tokenizer_config.json

@@ -0,0 +1,239 @@
+{
+  "add_bos_token": false,
+  "add_prefix_space": false,
+  "added_tokens_decoder": {
+    "151643": {
+      "content": "<|endoftext|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151644": {
+      "content": "<|im_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151645": {
+      "content": "<|im_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151646": {
+      "content": "<|object_ref_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151647": {
+      "content": "<|object_ref_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151648": {
+      "content": "<|box_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151649": {
+      "content": "<|box_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151650": {
+      "content": "<|quad_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151651": {
+      "content": "<|quad_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151652": {
+      "content": "<|vision_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151653": {
+      "content": "<|vision_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151654": {
+      "content": "<|vision_pad|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151655": {
+      "content": "<|image_pad|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151656": {
+      "content": "<|video_pad|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151657": {
+      "content": "<tool_call>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151658": {
+      "content": "</tool_call>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151659": {
+      "content": "<|fim_prefix|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151660": {
+      "content": "<|fim_middle|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151661": {
+      "content": "<|fim_suffix|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151662": {
+      "content": "<|fim_pad|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151663": {
+      "content": "<|repo_name|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151664": {
+      "content": "<|file_sep|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151665": {
+      "content": "<tool_response>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151666": {
+      "content": "</tool_response>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151667": {
+      "content": "<think>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151668": {
+      "content": "</think>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    }
+  },
+  "additional_special_tokens": [
+    "<|im_start|>",
+    "<|im_end|>",
+    "<|object_ref_start|>",
+    "<|object_ref_end|>",
+    "<|box_start|>",
+    "<|box_end|>",
+    "<|quad_start|>",
+    "<|quad_end|>",
+    "<|vision_start|>",
+    "<|vision_end|>",
+    "<|vision_pad|>",
+    "<|image_pad|>",
+    "<|video_pad|>"
+  ],
+  "bos_token": null,
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "<|im_end|>",
+  "errors": "replace",
+  "extra_special_tokens": {},
+  "model_max_length": 131072,
+  "pad_token": "<|endoftext|>",
+  "split_special_tokens": false,
+  "tokenizer_class": "Qwen2Tokenizer",
+  "unk_token": null
+}