AWQ W4A16_ASYM (group 128), llm-compressor, MTP preserved

.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 +1,132 @@
+---
+license: apache-2.0
+library_name: transformers
+pipeline_tag: image-text-to-text
+base_model: orcarouter/Qwen3.8-27B-Uncensored
+base_model_relation: quantized
+tags:
+  - awq
+  - w4a16
+  - int4
+  - llm-compressor
+  - compressed-tensors
+  - vllm
+  - qwen3_5
+  - multimodal
+  - mtp
+  - uncensored
+---
+
+# Qwen3.8-27B-Uncensored-AWQ-W4A16-ASYM
+
+4-битный AWQ-квант [orcarouter/Qwen3.8-27B-Uncensored](https://huggingface.co/orcarouter/Qwen3.8-27B-Uncensored),
+сделанный [llm-compressor](https://github.com/vllm-project/llm-compressor) в формате
+`compressed-tensors` (`pack-quantized`). Грузится в vLLM нативно.
+
+Квантование выполнено **с BF16-весов**, а не поверх FP8-чекпоинта — чтобы не накладывать
+INT4 на уже потерянную точность.
+
+## Что внутри
+
+| | |
+|---|---|
+| Схема | `W4A16_ASYM` (asymmetric, zero-point) |
+| Group size | 128 |
+| Стратегия | `group`, `type: int`, `num_bits: 4` |
+| Формат | `compressed-tensors` / `pack-quantized` |
+| Активации | BF16 (weight-only) |
+| Калибровка | `HuggingFaceH4/ultrachat_200k`, 256 сэмплов × 2048 токенов, через chat template |
+| Размер | **19 GB** (было 52 GB в BF16) |
+
+### Что квантовано, а что осталось в BF16
+
+Qwen3.8-27B — гибрид: 64 слоя по схеме `16 × (3 × Gated DeltaNet → 1 × Gated Attention)`,
+плюс vision-башня и MTP-голова.
+
+**В INT4:**
+- Gated Attention (слои 3, 7, …, 63): `q_proj`, `k_proj`, `v_proj`, `o_proj`
+- FFN всех 64 слоёв: `gate_proj`, `up_proj`, `down_proj`
+- Gated DeltaNet (остальные 48 слоёв): `in_proj_qkv`, `in_proj_z`, `out_proj`
+
+**Остались в BF16:**
+- `lm_head` и эмбеддинги
+- вся vision-башня (`model.visual.*`) — квантование визуального энкодера бьёт по качеству
+  распознавания сильнее, чем экономит память
+- `linear_attn.in_proj_a` / `in_proj_b` — чувствительные гейты SSM-ветки
+- вся MTP-голова (`mtp.*`) — чтобы не ломать спекулятивный декодинг
+
+> **Про MTP.** `transformers` не инстанцирует MTP-голову для `qwen3_5`, поэтому
+> llm-compressor её не видит и не сохраняет — наивный квант теряет MTP целиком.
+> Здесь тензоры `mtp.*` скопированы из исходного BF16-чекпоинта в готовый результат,
+> зарегистрированы в `model.safetensors.index.json` и внесены в
+> `quantization_config.ignore`. Спекулятивный декодинг сохранён.
+
+> **Про AWQ-mappings.** У гибридной архитектуры нет общего для всех слоёв набора проекций,
+> поэтому дефолтные маппинги AWQ не резолвятся. В `recipe.yaml` лежат кастомные: отдельный
+> маппинг для `input_layernorm` full-attention слоёв и отдельный — для DeltaNet-слоёв,
+> где в `balance_layers` обязаны быть **все** потребители выхода нормы, включая
+> неквантуемые `in_proj_a` / `in_proj_b`, иначе сглаживание ломает математику.
+
+## Использование (vLLM)
+
+```bash
+vllm serve Ar4ikov/Qwen3.8-27B-Uncensored-AWQ-W4A16-ASYM \
+  --max-model-len 262144 \
+  --tensor-parallel-size 2
+```
+
+## Использование (transformers)
+
+```python
+from transformers import AutoModelForImageTextToText, AutoProcessor
+
+model = AutoModelForImageTextToText.from_pretrained(
+    "Ar4ikov/Qwen3.8-27B-Uncensored-AWQ-W4A16-ASYM", dtype="auto", device_map="auto"
+)
+processor = AutoProcessor.from_pretrained("Ar4ikov/Qwen3.8-27B-Uncensored-AWQ-W4A16-ASYM")
+```
+
+> transformers при загрузке **распаковывает** веса обратно в BF16, так что экономии VRAM
+> здесь не будет. Реальная экономия видна только в рантаймах с нативными INT4-ядрами
+> (vLLM, SGLang): ожидаемо ≈ размер файла + KV-кэш.
+
+## Параметры сэмплирования
+
+Наследуются от базовой модели:
+
+- **Thinking mode:** `temperature=1.0`, `top_p=0.95`, `top_k=20`, `min_p=0.0`, `presence_penalty=0.0`
+- **Non-thinking:** `temperature=0.7`, `top_p=0.80`, `top_k=20`, `min_p=0.0`, `presence_penalty=1.5`
+
+Поддерживается `reasoning_effort` (`low` / `medium` / `xhigh`), контекст 262 144 токена нативно.
+
+## Воспроизводимость
+
+В репозитории лежит `recipe.yaml`, сгене��ированный llm-compressor. Ключевая часть:
+
+```python
+from llmcompressor.modifiers.awq import AWQModifier
+
+recipe = [AWQModifier(
+    targets=["Linear"],
+    scheme="W4A16_ASYM",
+    ignore=[
+        "re:.*lm_head",
+        r"re:model\.visual\..*",
+        r"re:.*\.linear_attn\.in_proj_a",
+        r"re:.*\.linear_attn\.in_proj_b",
+        r"re:mtp\..*",
+    ],
+    mappings=MAPPINGS,   # см. recipe.yaml
+)]
+```
+
+## Оговорки
+
+- Квантовано 400 Linear-слоёв (64×3 MLP + 16×4 full-attention + 48×3 DeltaNet);
+  400 тензоров `weight_packed` (I32-упаковка INT4) с `weight_scale` и `weight_zero_point`.
+- Квант не проходил бенчмарк-валидацию — проверена только связность генерации.
+  Для продакшена прогоните свои эвалы.
+- Vision-часть не квантовалась, поэтому качество на изображениях должно быть близко
+  к исходной модели; деградация ожидается прежде всего в текстовой части.
+- Базовая модель — uncensored-файнтюн; квантование не добавляет и не снимает никаких
+  ограничений, поведение наследуется от `orcarouter/Qwen3.8-27B-Uncensored`.

chat_template.jinja

@@ -0,0 +1,170 @@
+{%- set image_count = namespace(value=0) %}
+{%- set video_count = namespace(value=0) %}
+{%- macro render_content(content, do_vision_count, is_system_content=false) %}
+    {%- if content is string %}
+        {{- content }}
+    {%- elif content is iterable and content is not mapping %}
+        {%- for item in content %}
+            {%- if 'image' in item or 'image_url' in item or item.type == 'image' %}
+                {%- if is_system_content %}
+                    {{- raise_exception('System message cannot contain images.') }}
+                {%- endif %}
+                {%- if do_vision_count %}
+                    {%- set image_count.value = image_count.value + 1 %}
+                {%- endif %}
+                {%- if add_vision_id %}
+                    {{- 'Picture ' ~ image_count.value ~ ': ' }}
+                {%- endif %}
+                {{- '<|vision_start|><|image_pad|><|vision_end|>' }}
+            {%- elif 'video' in item or item.type == 'video' %}
+                {%- if is_system_content %}
+                    {{- raise_exception('System message cannot contain videos.') }}
+                {%- endif %}
+                {%- if do_vision_count %}
+                    {%- set video_count.value = video_count.value + 1 %}
+                {%- endif %}
+                {%- if add_vision_id %}
+                    {{- 'Video ' ~ video_count.value ~ ': ' }}
+                {%- endif %}
+                {{- '<|vision_start|><|video_pad|><|vision_end|>' }}
+            {%- elif 'text' in item %}
+                {{- item.text }}
+            {%- else %}
+                {{- raise_exception('Unexpected item type in content.') }}
+            {%- endif %}
+        {%- endfor %}
+    {%- elif content is none or content is undefined %}
+        {{- '' }}
+    {%- else %}
+        {{- raise_exception('Unexpected content type.') }}
+    {%- endif %}
+{%- endmacro %}
+{%- if not messages %}
+    {{- raise_exception('No messages provided.') }}
+{%- endif %}
+{%- set reasoning_instructions = '' %}
+{%- if enable_thinking is undefined or enable_thinking is true %}
+    {%- set resolved_reasoning_effort = reasoning_effort|default('xhigh') %}
+    {%- if resolved_reasoning_effort not in ('xhigh', 'medium', 'low') %}
+        {{- raise_exception('Unexpected reasoning effort ' ~ reasoning_effort ~ '. Supported types are xhigh (default), medium, and low.') }}
+    {%- endif %}
+    {%- if resolved_reasoning_effort == 'xhigh' %}
+        {%- set reasoning_instructions = 'Reasoning effort is set to xhigh. Please think carefully through the task, validate key assumptions, consider plausible alternatives, and prioritize correctness, consistency, and clarity in the final answer.' %}
+    {%- elif resolved_reasoning_effort == 'low' %}
+        {%- set reasoning_instructions = 'Reasoning effort is set to low. Keep your thinking brief and focused, moving directly to the conclusion without unnecessary elaboration.' %}
+    {%- endif %}
+{%- endif %}
+{%- if tools and tools is iterable and tools is not mapping %}
+    {{- '<|im_start|>system\n' }}
+    {%- if reasoning_instructions %}
+        {{- reasoning_instructions + '\n\n' }}
+    {%- endif %}
+    {{- "# Tools\n\nYou have access to the following functions:\n\n<tools>" }}
+    {%- for tool in tools %}
+        {{- "\n" }}
+        {{- tool | tojson }}
+    {%- endfor %}
+    {{- "\n</tools>" }}
+    {{- '\n\nIf you choose to call a function ONLY reply in the following format with NO suffix:\n\n<tool_call>\n<function=example_function_name>\n<parameter=example_parameter_1>\nvalue_1\n</parameter>\n<parameter=example_parameter_2>\nThis is the value for the second parameter\nthat can span\nmultiple lines\n</parameter>\n</function>\n</tool_call>\n\n<IMPORTANT>\nReminder:\n- Function calls MUST follow the specified format: an inner <function=...></function> block must be nested within <tool_call></tool_call> XML tags\n- Required parameters MUST be specified\n- You may provide optional reasoning for your function call in natural language BEFORE the function call, but NOT after\n- If there is no function call available, answer the question like normal with your current knowledge and do not tell the user about function calls\n</IMPORTANT>' }}
+    {%- if messages[0].role == 'system' %}
+        {%- set content = render_content(messages[0].content, false, true)|trim %}
+        {%- if content %}
+            {{- '\n\n' + content }}
+        {%- endif %}
+    {%- endif %}
+    {{- '<|im_end|>\n' }}
+{%- else %}
+    {%- if messages[0].role == 'system' %}
+        {%- set content = render_content(messages[0].content, false, true)|trim %}
+        {%- if content %}
+            {{- '<|im_start|>system\n' + (reasoning_instructions + '\n\n' if reasoning_instructions else '')  + content + '<|im_end|>\n' }}
+        {%- elif reasoning_instructions %}
+            {{- '<|im_start|>system\n' + reasoning_instructions + '<|im_end|>\n' }}
+        {%- endif %}
+    {%- elif reasoning_instructions %}
+        {{- '<|im_start|>system\n' + reasoning_instructions + '<|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" %}
+        {%- set content = render_content(message.content, false)|trim %}
+        {%- if not(content.startswith('<tool_response>') and content.endswith('</tool_response>')) %}
+            {%- set ns.multi_step_tool = false %}
+            {%- set ns.last_query_index = index %}
+        {%- endif %}
+    {%- endif %}
+{%- endfor %}
+{%- if ns.multi_step_tool %}
+    {{- raise_exception('No user query found in messages.') }}
+{%- endif %}
+{%- for message in messages %}
+    {%- set content = render_content(message.content, true)|trim %}
+    {%- if message.role == "system" %}
+        {%- if not loop.first %}
+            {{- raise_exception('System message must be at the beginning.') }}
+        {%- endif %}
+    {%- elif message.role == "user" %}
+        {{- '<|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 %}
+        {%- endif %}
+        {%- set reasoning_content = reasoning_content|trim %}
+        {%- if preserve_thinking is undefined or preserve_thinking is true or loop.index0 > ns.last_query_index %}
+            {{- '<|im_start|>' + message.role + '\n<think>\n' + reasoning_content + '\n</think>\n\n' + content }}
+        {%- else %}
+            {{- '<|im_start|>' + message.role + '\n' + content }}
+        {%- endif %}
+        {%- if message.tool_calls and message.tool_calls is iterable and message.tool_calls is not mapping %}
+            {%- for tool_call in message.tool_calls %}
+                {%- if tool_call.function is defined %}
+                    {%- set tool_call = tool_call.function %}
+                {%- endif %}
+                {%- if loop.first %}
+                    {%- if content|trim %}
+                        {{- '\n\n<tool_call>\n<function=' + tool_call.name + '>\n' }}
+                    {%- else %}
+                        {{- '<tool_call>\n<function=' + tool_call.name + '>\n' }}
+                    {%- endif %}
+                {%- else %}
+                    {{- '\n<tool_call>\n<function=' + tool_call.name + '>\n' }}
+                {%- endif %}
+                {%- if tool_call.arguments is defined and tool_call.arguments != '' %}
+                    {%- for args_name, args_value in tool_call.arguments|items %}
+                        {{- '<parameter=' + args_name + '>\n' }}
+                        {%- set args_value = args_value | string if args_value is string else args_value | tojson | safe %}
+                        {{- args_value }}
+                        {{- '\n</parameter>\n' }}
+                    {%- endfor %}
+                {%- endif %}
+                {{- '</function>\n</tool_call>' }}
+            {%- endfor %}
+        {%- endif %}
+        {{- '<|im_end|>\n' }}
+    {%- elif message.role == "tool" %}
+        {%- if loop.previtem and loop.previtem.role != "tool" %}
+            {{- '<|im_start|>user' }}
+        {%- endif %}
+        {{- '\n<tool_response>\n' }}
+        {{- content }}
+        {{- '\n</tool_response>' }}
+        {%- if not loop.last and loop.nextitem.role != "tool" %}
+            {{- '<|im_end|>\n' }}
+        {%- elif loop.last %}
+            {{- '<|im_end|>\n' }}
+        {%- endif %}
+    {%- else %}
+        {{- raise_exception('Unexpected message role.') }}
+    {%- 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' }}
+    {%- else %}
+        {{- '<think>\n' }}
+    {%- endif %}
+{%- endif %}

config.json

@@ -0,0 +1,489 @@
+{
+  "architectures": [
+    "Qwen3_5ForConditionalGeneration"
+  ],
+  "dtype": "bfloat16",
+  "image_token_id": 248056,
+  "language_model_only": false,
+  "model_type": "qwen3_5",
+  "quantization_config": {
+    "config_groups": {
+      "group_0": {
+        "format": "pack-quantized",
+        "input_activations": null,
+        "output_activations": null,
+        "targets": [
+          "Linear"
+        ],
+        "weights": {
+          "actorder": null,
+          "block_structure": null,
+          "dynamic": false,
+          "group_size": 128,
+          "num_bits": 4,
+          "observer": "memoryless_minmax",
+          "observer_kwargs": {},
+          "scale_dtype": null,
+          "strategy": "group",
+          "symmetric": false,
+          "type": "int",
+          "zp_dtype": "torch.int8"
+        }
+      }
+    },
+    "format": "pack-quantized",
+    "global_compression_ratio": null,
+    "ignore": [
+      "model.visual.blocks.0.attn.qkv",
+      "model.visual.blocks.0.attn.proj",
+      "model.visual.blocks.0.mlp.linear_fc1",
+      "model.visual.blocks.0.mlp.linear_fc2",
+      "model.visual.blocks.1.attn.qkv",
+      "model.visual.blocks.1.attn.proj",
+      "model.visual.blocks.1.mlp.linear_fc1",
+      "model.visual.blocks.1.mlp.linear_fc2",
+      "model.visual.blocks.2.attn.qkv",
+      "model.visual.blocks.2.attn.proj",
+      "model.visual.blocks.2.mlp.linear_fc1",
+      "model.visual.blocks.2.mlp.linear_fc2",
+      "model.visual.blocks.3.attn.qkv",
+      "model.visual.blocks.3.attn.proj",
+      "model.visual.blocks.3.mlp.linear_fc1",
+      "model.visual.blocks.3.mlp.linear_fc2",
+      "model.visual.blocks.4.attn.qkv",
+      "model.visual.blocks.4.attn.proj",
+      "model.visual.blocks.4.mlp.linear_fc1",
+      "model.visual.blocks.4.mlp.linear_fc2",
+      "model.visual.blocks.5.attn.qkv",
+      "model.visual.blocks.5.attn.proj",
+      "model.visual.blocks.5.mlp.linear_fc1",
+      "model.visual.blocks.5.mlp.linear_fc2",
+      "model.visual.blocks.6.attn.qkv",
+      "model.visual.blocks.6.attn.proj",
+      "model.visual.blocks.6.mlp.linear_fc1",
+      "model.visual.blocks.6.mlp.linear_fc2",
+      "model.visual.blocks.7.attn.qkv",
+      "model.visual.blocks.7.attn.proj",
+      "model.visual.blocks.7.mlp.linear_fc1",
+      "model.visual.blocks.7.mlp.linear_fc2",
+      "model.visual.blocks.8.attn.qkv",
+      "model.visual.blocks.8.attn.proj",
+      "model.visual.blocks.8.mlp.linear_fc1",
+      "model.visual.blocks.8.mlp.linear_fc2",
+      "model.visual.blocks.9.attn.qkv",
+      "model.visual.blocks.9.attn.proj",
+      "model.visual.blocks.9.mlp.linear_fc1",
+      "model.visual.blocks.9.mlp.linear_fc2",
+      "model.visual.blocks.10.attn.qkv",
+      "model.visual.blocks.10.attn.proj",
+      "model.visual.blocks.10.mlp.linear_fc1",
+      "model.visual.blocks.10.mlp.linear_fc2",
+      "model.visual.blocks.11.attn.qkv",
+      "model.visual.blocks.11.attn.proj",
+      "model.visual.blocks.11.mlp.linear_fc1",
+      "model.visual.blocks.11.mlp.linear_fc2",
+      "model.visual.blocks.12.attn.qkv",
+      "model.visual.blocks.12.attn.proj",
+      "model.visual.blocks.12.mlp.linear_fc1",
+      "model.visual.blocks.12.mlp.linear_fc2",
+      "model.visual.blocks.13.attn.qkv",
+      "model.visual.blocks.13.attn.proj",
+      "model.visual.blocks.13.mlp.linear_fc1",
+      "model.visual.blocks.13.mlp.linear_fc2",
+      "model.visual.blocks.14.attn.qkv",
+      "model.visual.blocks.14.attn.proj",
+      "model.visual.blocks.14.mlp.linear_fc1",
+      "model.visual.blocks.14.mlp.linear_fc2",
+      "model.visual.blocks.15.attn.qkv",
+      "model.visual.blocks.15.attn.proj",
+      "model.visual.blocks.15.mlp.linear_fc1",
+      "model.visual.blocks.15.mlp.linear_fc2",
+      "model.visual.blocks.16.attn.qkv",
+      "model.visual.blocks.16.attn.proj",
+      "model.visual.blocks.16.mlp.linear_fc1",
+      "model.visual.blocks.16.mlp.linear_fc2",
+      "model.visual.blocks.17.attn.qkv",
+      "model.visual.blocks.17.attn.proj",
+      "model.visual.blocks.17.mlp.linear_fc1",
+      "model.visual.blocks.17.mlp.linear_fc2",
+      "model.visual.blocks.18.attn.qkv",
+      "model.visual.blocks.18.attn.proj",
+      "model.visual.blocks.18.mlp.linear_fc1",
+      "model.visual.blocks.18.mlp.linear_fc2",
+      "model.visual.blocks.19.attn.qkv",
+      "model.visual.blocks.19.attn.proj",
+      "model.visual.blocks.19.mlp.linear_fc1",
+      "model.visual.blocks.19.mlp.linear_fc2",
+      "model.visual.blocks.20.attn.qkv",
+      "model.visual.blocks.20.attn.proj",
+      "model.visual.blocks.20.mlp.linear_fc1",
+      "model.visual.blocks.20.mlp.linear_fc2",
+      "model.visual.blocks.21.attn.qkv",
+      "model.visual.blocks.21.attn.proj",
+      "model.visual.blocks.21.mlp.linear_fc1",
+      "model.visual.blocks.21.mlp.linear_fc2",
+      "model.visual.blocks.22.attn.qkv",
+      "model.visual.blocks.22.attn.proj",
+      "model.visual.blocks.22.mlp.linear_fc1",
+      "model.visual.blocks.22.mlp.linear_fc2",
+      "model.visual.blocks.23.attn.qkv",
+      "model.visual.blocks.23.attn.proj",
+      "model.visual.blocks.23.mlp.linear_fc1",
+      "model.visual.blocks.23.mlp.linear_fc2",
+      "model.visual.blocks.24.attn.qkv",
+      "model.visual.blocks.24.attn.proj",
+      "model.visual.blocks.24.mlp.linear_fc1",
+      "model.visual.blocks.24.mlp.linear_fc2",
+      "model.visual.blocks.25.attn.qkv",
+      "model.visual.blocks.25.attn.proj",
+      "model.visual.blocks.25.mlp.linear_fc1",
+      "model.visual.blocks.25.mlp.linear_fc2",
+      "model.visual.blocks.26.attn.qkv",
+      "model.visual.blocks.26.attn.proj",
+      "model.visual.blocks.26.mlp.linear_fc1",
+      "model.visual.blocks.26.mlp.linear_fc2",
+      "model.visual.merger.linear_fc1",
+      "model.visual.merger.linear_fc2",
+      "model.language_model.layers.0.linear_attn",
+      "model.language_model.layers.0.linear_attn.norm",
+      "model.language_model.layers.0.linear_attn.in_proj_b",
+      "model.language_model.layers.0.linear_attn.in_proj_a",
+      "model.language_model.layers.1.linear_attn",
+      "model.language_model.layers.1.linear_attn.norm",
+      "model.language_model.layers.1.linear_attn.in_proj_b",
+      "model.language_model.layers.1.linear_attn.in_proj_a",
+      "model.language_model.layers.2.linear_attn",
+      "model.language_model.layers.2.linear_attn.norm",
+      "model.language_model.layers.2.linear_attn.in_proj_b",
+      "model.language_model.layers.2.linear_attn.in_proj_a",
+      "model.language_model.layers.4.linear_attn",
+      "model.language_model.layers.4.linear_attn.norm",
+      "model.language_model.layers.4.linear_attn.in_proj_b",
+      "model.language_model.layers.4.linear_attn.in_proj_a",
+      "model.language_model.layers.5.linear_attn",
+      "model.language_model.layers.5.linear_attn.norm",
+      "model.language_model.layers.5.linear_attn.in_proj_b",
+      "model.language_model.layers.5.linear_attn.in_proj_a",
+      "model.language_model.layers.6.linear_attn",
+      "model.language_model.layers.6.linear_attn.norm",
+      "model.language_model.layers.6.linear_attn.in_proj_b",
+      "model.language_model.layers.6.linear_attn.in_proj_a",
+      "model.language_model.layers.8.linear_attn",
+      "model.language_model.layers.8.linear_attn.norm",
+      "model.language_model.layers.8.linear_attn.in_proj_b",
+      "model.language_model.layers.8.linear_attn.in_proj_a",
+      "model.language_model.layers.9.linear_attn",
+      "model.language_model.layers.9.linear_attn.norm",
+      "model.language_model.layers.9.linear_attn.in_proj_b",
+      "model.language_model.layers.9.linear_attn.in_proj_a",
+      "model.language_model.layers.10.linear_attn",
+      "model.language_model.layers.10.linear_attn.norm",
+      "model.language_model.layers.10.linear_attn.in_proj_b",
+      "model.language_model.layers.10.linear_attn.in_proj_a",
+      "model.language_model.layers.12.linear_attn",
+      "model.language_model.layers.12.linear_attn.norm",
+      "model.language_model.layers.12.linear_attn.in_proj_b",
+      "model.language_model.layers.12.linear_attn.in_proj_a",
+      "model.language_model.layers.13.linear_attn",
+      "model.language_model.layers.13.linear_attn.norm",
+      "model.language_model.layers.13.linear_attn.in_proj_b",
+      "model.language_model.layers.13.linear_attn.in_proj_a",
+      "model.language_model.layers.14.linear_attn",
+      "model.language_model.layers.14.linear_attn.norm",
+      "model.language_model.layers.14.linear_attn.in_proj_b",
+      "model.language_model.layers.14.linear_attn.in_proj_a",
+      "model.language_model.layers.16.linear_attn",
+      "model.language_model.layers.16.linear_attn.norm",
+      "model.language_model.layers.16.linear_attn.in_proj_b",
+      "model.language_model.layers.16.linear_attn.in_proj_a",
+      "model.language_model.layers.17.linear_attn",
+      "model.language_model.layers.17.linear_attn.norm",
+      "model.language_model.layers.17.linear_attn.in_proj_b",
+      "model.language_model.layers.17.linear_attn.in_proj_a",
+      "model.language_model.layers.18.linear_attn",
+      "model.language_model.layers.18.linear_attn.norm",
+      "model.language_model.layers.18.linear_attn.in_proj_b",
+      "model.language_model.layers.18.linear_attn.in_proj_a",
+      "model.language_model.layers.20.linear_attn",
+      "model.language_model.layers.20.linear_attn.norm",
+      "model.language_model.layers.20.linear_attn.in_proj_b",
+      "model.language_model.layers.20.linear_attn.in_proj_a",
+      "model.language_model.layers.21.linear_attn",
+      "model.language_model.layers.21.linear_attn.norm",
+      "model.language_model.layers.21.linear_attn.in_proj_b",
+      "model.language_model.layers.21.linear_attn.in_proj_a",
+      "model.language_model.layers.22.linear_attn",
+      "model.language_model.layers.22.linear_attn.norm",
+      "model.language_model.layers.22.linear_attn.in_proj_b",
+      "model.language_model.layers.22.linear_attn.in_proj_a",
+      "model.language_model.layers.24.linear_attn",
+      "model.language_model.layers.24.linear_attn.norm",
+      "model.language_model.layers.24.linear_attn.in_proj_b",
+      "model.language_model.layers.24.linear_attn.in_proj_a",
+      "model.language_model.layers.25.linear_attn",
+      "model.language_model.layers.25.linear_attn.norm",
+      "model.language_model.layers.25.linear_attn.in_proj_b",
+      "model.language_model.layers.25.linear_attn.in_proj_a",
+      "model.language_model.layers.26.linear_attn",
+      "model.language_model.layers.26.linear_attn.norm",
+      "model.language_model.layers.26.linear_attn.in_proj_b",
+      "model.language_model.layers.26.linear_attn.in_proj_a",
+      "model.language_model.layers.28.linear_attn",
+      "model.language_model.layers.28.linear_attn.norm",
+      "model.language_model.layers.28.linear_attn.in_proj_b",
+      "model.language_model.layers.28.linear_attn.in_proj_a",
+      "model.language_model.layers.29.linear_attn",
+      "model.language_model.layers.29.linear_attn.norm",
+      "model.language_model.layers.29.linear_attn.in_proj_b",
+      "model.language_model.layers.29.linear_attn.in_proj_a",
+      "model.language_model.layers.30.linear_attn",
+      "model.language_model.layers.30.linear_attn.norm",
+      "model.language_model.layers.30.linear_attn.in_proj_b",
+      "model.language_model.layers.30.linear_attn.in_proj_a",
+      "model.language_model.layers.32.linear_attn",
+      "model.language_model.layers.32.linear_attn.norm",
+      "model.language_model.layers.32.linear_attn.in_proj_b",
+      "model.language_model.layers.32.linear_attn.in_proj_a",
+      "model.language_model.layers.33.linear_attn",
+      "model.language_model.layers.33.linear_attn.norm",
+      "model.language_model.layers.33.linear_attn.in_proj_b",
+      "model.language_model.layers.33.linear_attn.in_proj_a",
+      "model.language_model.layers.34.linear_attn",
+      "model.language_model.layers.34.linear_attn.norm",
+      "model.language_model.layers.34.linear_attn.in_proj_b",
+      "model.language_model.layers.34.linear_attn.in_proj_a",
+      "model.language_model.layers.36.linear_attn",
+      "model.language_model.layers.36.linear_attn.norm",
+      "model.language_model.layers.36.linear_attn.in_proj_b",
+      "model.language_model.layers.36.linear_attn.in_proj_a",
+      "model.language_model.layers.37.linear_attn",
+      "model.language_model.layers.37.linear_attn.norm",
+      "model.language_model.layers.37.linear_attn.in_proj_b",
+      "model.language_model.layers.37.linear_attn.in_proj_a",
+      "model.language_model.layers.38.linear_attn",
+      "model.language_model.layers.38.linear_attn.norm",
+      "model.language_model.layers.38.linear_attn.in_proj_b",
+      "model.language_model.layers.38.linear_attn.in_proj_a",
+      "model.language_model.layers.40.linear_attn",
+      "model.language_model.layers.40.linear_attn.norm",
+      "model.language_model.layers.40.linear_attn.in_proj_b",
+      "model.language_model.layers.40.linear_attn.in_proj_a",
+      "model.language_model.layers.41.linear_attn",
+      "model.language_model.layers.41.linear_attn.norm",
+      "model.language_model.layers.41.linear_attn.in_proj_b",
+      "model.language_model.layers.41.linear_attn.in_proj_a",
+      "model.language_model.layers.42.linear_attn",
+      "model.language_model.layers.42.linear_attn.norm",
+      "model.language_model.layers.42.linear_attn.in_proj_b",
+      "model.language_model.layers.42.linear_attn.in_proj_a",
+      "model.language_model.layers.44.linear_attn",
+      "model.language_model.layers.44.linear_attn.norm",
+      "model.language_model.layers.44.linear_attn.in_proj_b",
+      "model.language_model.layers.44.linear_attn.in_proj_a",
+      "model.language_model.layers.45.linear_attn",
+      "model.language_model.layers.45.linear_attn.norm",
+      "model.language_model.layers.45.linear_attn.in_proj_b",
+      "model.language_model.layers.45.linear_attn.in_proj_a",
+      "model.language_model.layers.46.linear_attn",
+      "model.language_model.layers.46.linear_attn.norm",
+      "model.language_model.layers.46.linear_attn.in_proj_b",
+      "model.language_model.layers.46.linear_attn.in_proj_a",
+      "model.language_model.layers.48.linear_attn",
+      "model.language_model.layers.48.linear_attn.norm",
+      "model.language_model.layers.48.linear_attn.in_proj_b",
+      "model.language_model.layers.48.linear_attn.in_proj_a",
+      "model.language_model.layers.49.linear_attn",
+      "model.language_model.layers.49.linear_attn.norm",
+      "model.language_model.layers.49.linear_attn.in_proj_b",
+      "model.language_model.layers.49.linear_attn.in_proj_a",
+      "model.language_model.layers.50.linear_attn",
+      "model.language_model.layers.50.linear_attn.norm",
+      "model.language_model.layers.50.linear_attn.in_proj_b",
+      "model.language_model.layers.50.linear_attn.in_proj_a",
+      "model.language_model.layers.52.linear_attn",
+      "model.language_model.layers.52.linear_attn.norm",
+      "model.language_model.layers.52.linear_attn.in_proj_b",
+      "model.language_model.layers.52.linear_attn.in_proj_a",
+      "model.language_model.layers.53.linear_attn",
+      "model.language_model.layers.53.linear_attn.norm",
+      "model.language_model.layers.53.linear_attn.in_proj_b",
+      "model.language_model.layers.53.linear_attn.in_proj_a",
+      "model.language_model.layers.54.linear_attn",
+      "model.language_model.layers.54.linear_attn.norm",
+      "model.language_model.layers.54.linear_attn.in_proj_b",
+      "model.language_model.layers.54.linear_attn.in_proj_a",
+      "model.language_model.layers.56.linear_attn",
+      "model.language_model.layers.56.linear_attn.norm",
+      "model.language_model.layers.56.linear_attn.in_proj_b",
+      "model.language_model.layers.56.linear_attn.in_proj_a",
+      "model.language_model.layers.57.linear_attn",
+      "model.language_model.layers.57.linear_attn.norm",
+      "model.language_model.layers.57.linear_attn.in_proj_b",
+      "model.language_model.layers.57.linear_attn.in_proj_a",
+      "model.language_model.layers.58.linear_attn",
+      "model.language_model.layers.58.linear_attn.norm",
+      "model.language_model.layers.58.linear_attn.in_proj_b",
+      "model.language_model.layers.58.linear_attn.in_proj_a",
+      "model.language_model.layers.60.linear_attn",
+      "model.language_model.layers.60.linear_attn.norm",
+      "model.language_model.layers.60.linear_attn.in_proj_b",
+      "model.language_model.layers.60.linear_attn.in_proj_a",
+      "model.language_model.layers.61.linear_attn",
+      "model.language_model.layers.61.linear_attn.norm",
+      "model.language_model.layers.61.linear_attn.in_proj_b",
+      "model.language_model.layers.61.linear_attn.in_proj_a",
+      "model.language_model.layers.62.linear_attn",
+      "model.language_model.layers.62.linear_attn.norm",
+      "model.language_model.layers.62.linear_attn.in_proj_b",
+      "model.language_model.layers.62.linear_attn.in_proj_a",
+      "lm_head",
+      "mtp.fc",
+      "mtp.layers.0.mlp.down_proj",
+      "mtp.layers.0.mlp.gate_proj",
+      "mtp.layers.0.mlp.up_proj",
+      "mtp.layers.0.self_attn.k_proj",
+      "mtp.layers.0.self_attn.o_proj",
+      "mtp.layers.0.self_attn.q_proj",
+      "mtp.layers.0.self_attn.v_proj"
+    ],
+    "kv_cache_scheme": null,
+    "quant_method": "compressed-tensors",
+    "quantization_status": "compressed",
+    "sparsity_config": {},
+    "transform_config": {},
+    "version": "0.18.0"
+  },
+  "text_config": {
+    "attention_bias": false,
+    "attention_dropout": 0.0,
+    "attn_output_gate": true,
+    "bos_token_id": 248044,
+    "dtype": "bfloat16",
+    "eos_token_id": 248044,
+    "full_attention_interval": 4,
+    "head_dim": 256,
+    "hidden_act": "silu",
+    "hidden_size": 5120,
+    "initializer_range": 0.02,
+    "intermediate_size": 17408,
+    "layer_types": [
+      "linear_attention",
+      "linear_attention",
+      "linear_attention",
+      "full_attention",
+      "linear_attention",
+      "linear_attention",
+      "linear_attention",
+      "full_attention",
+      "linear_attention",
+      "linear_attention",
+      "linear_attention",
+      "full_attention",
+      "linear_attention",
+      "linear_attention",
+      "linear_attention",
+      "full_attention",
+      "linear_attention",
+      "linear_attention",
+      "linear_attention",
+      "full_attention",
+      "linear_attention",
+      "linear_attention",
+      "linear_attention",
+      "full_attention",
+      "linear_attention",
+      "linear_attention",
+      "linear_attention",
+      "full_attention",
+      "linear_attention",
+      "linear_attention",
+      "linear_attention",
+      "full_attention",
+      "linear_attention",
+      "linear_attention",
+      "linear_attention",
+      "full_attention",
+      "linear_attention",
+      "linear_attention",
+      "linear_attention",
+      "full_attention",
+      "linear_attention",
+      "linear_attention",
+      "linear_attention",
+      "full_attention",
+      "linear_attention",
+      "linear_attention",
+      "linear_attention",
+      "full_attention",
+      "linear_attention",
+      "linear_attention",
+      "linear_attention",
+      "full_attention",
+      "linear_attention",
+      "linear_attention",
+      "linear_attention",
+      "full_attention",
+      "linear_attention",
+      "linear_attention",
+      "linear_attention",
+      "full_attention",
+      "linear_attention",
+      "linear_attention",
+      "linear_attention",
+      "full_attention"
+    ],
+    "linear_conv_kernel_dim": 4,
+    "linear_key_head_dim": 128,
+    "linear_num_key_heads": 16,
+    "linear_num_value_heads": 48,
+    "linear_value_head_dim": 128,
+    "mamba_ssm_dtype": "float32",
+    "max_position_embeddings": 262144,
+    "model_type": "qwen3_5_text",
+    "mtp_num_hidden_layers": 1,
+    "mtp_use_dedicated_embeddings": false,
+    "num_attention_heads": 24,
+    "num_hidden_layers": 64,
+    "num_key_value_heads": 4,
+    "output_gate_type": "swish",
+    "pad_token_id": null,
+    "partial_rotary_factor": 0.25,
+    "rms_norm_eps": 1e-06,
+    "rope_parameters": {
+      "mrope_interleaved": true,
+      "mrope_section": [
+        11,
+        11,
+        10
+      ],
+      "partial_rotary_factor": 0.25,
+      "rope_theta": 10000000,
+      "rope_type": "default"
+    },
+    "tie_word_embeddings": false,
+    "use_cache": true,
+    "vocab_size": 248320
+  },
+  "tie_word_embeddings": false,
+  "transformers_version": "5.14.1",
+  "video_token_id": 248057,
+  "vision_config": {
+    "deepstack_visual_indexes": [],
+    "depth": 27,
+    "dtype": "bfloat16",
+    "hidden_act": "gelu_pytorch_tanh",
+    "hidden_size": 1152,
+    "in_channels": 3,
+    "initializer_range": 0.02,
+    "intermediate_size": 4304,
+    "model_type": "qwen3_5_vision",
+    "num_heads": 16,
+    "num_position_embeddings": 2304,
+    "out_hidden_size": 5120,
+    "patch_size": 16,
+    "spatial_merge_size": 2,
+    "temporal_patch_size": 2
+  },
+  "vision_end_token_id": 248054,
+  "vision_start_token_id": 248053
+}

generation_config.json

@@ -0,0 +1,13 @@
+{
+  "bos_token_id": 248044,
+  "do_sample": true,
+  "eos_token_id": [
+    248046,
+    248044
+  ],
+  "pad_token_id": 248044,
+  "temperature": 1.0,
+  "top_k": 20,
+  "top_p": 0.95,
+  "transformers_version": "5.14.1"
+}

model-mtp.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d869b1dfb497f3ca5af94e77417447045c1468716314c224c83ee237ff01293a
+size 849400424

model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1daf05d5ddb4e968d8e86b4c00ed60f28fb07b268fdfcb19022b17861a09d812
+size 18698467264

preprocessor_config.json

@@ -0,0 +1,21 @@
+{
+    "size": {
+        "longest_edge": 16777216,
+        "shortest_edge": 65536
+    },
+    "patch_size": 16,
+    "temporal_patch_size": 2,
+    "merge_size": 2,
+    "image_mean": [
+        0.5,
+        0.5,
+        0.5
+    ],
+    "image_std": [
+        0.5,
+        0.5,
+        0.5
+    ],
+    "processor_class": "Qwen3VLProcessor",
+    "image_processor_type": "Qwen2VLImageProcessorFast"
+}

processor_config.json

@@ -0,0 +1,60 @@
+{
+  "image_processor": {
+    "do_convert_rgb": true,
+    "do_normalize": true,
+    "do_rescale": true,
+    "do_resize": true,
+    "image_mean": [
+      0.5,
+      0.5,
+      0.5
+    ],
+    "image_processor_type": "Qwen2VLImageProcessor",
+    "image_std": [
+      0.5,
+      0.5,
+      0.5
+    ],
+    "merge_size": 2,
+    "patch_size": 16,
+    "resample": 3,
+    "rescale_factor": 0.00392156862745098,
+    "size": {
+      "longest_edge": 16777216,
+      "shortest_edge": 65536
+    },
+    "temporal_patch_size": 2
+  },
+  "processor_class": "Qwen3VLProcessor",
+  "video_processor": {
+    "do_convert_rgb": true,
+    "do_normalize": true,
+    "do_rescale": true,
+    "do_resize": true,
+    "do_sample_frames": true,
+    "fps": 2,
+    "image_mean": [
+      0.5,
+      0.5,
+      0.5
+    ],
+    "image_std": [
+      0.5,
+      0.5,
+      0.5
+    ],
+    "max_frames": 768,
+    "merge_size": 2,
+    "min_frames": 4,
+    "patch_size": 16,
+    "resample": 3,
+    "rescale_factor": 0.00392156862745098,
+    "return_metadata": false,
+    "size": {
+      "longest_edge": 25165824,
+      "shortest_edge": 4096
+    },
+    "temporal_patch_size": 2,
+    "video_processor_type": "Qwen3VLVideoProcessor"
+  }
+}

recipe.yaml

@@ -0,0 +1,28 @@
+default_stage:
+  default_modifiers:
+    AWQModifier:
+      requires_calibration_data: true
+      mappings:
+      - smooth_layer: re:.*layers\.(3|7|11|15|19|23|27|31|35|39|43|47|51|55|59|63)\.input_layernorm$
+        balance_layers: ['re:.*self_attn.q_proj$', 're:.*self_attn.k_proj$', 're:.*self_attn.v_proj$']
+        activation_hook_target: null
+      - smooth_layer: re:.*layers\.(0|1|2|4|5|6|8|9|10|12|13|14|16|17|18|20|21|22|24|25|26|28|29|30|32|33|34|36|37|38|40|41|42|44|45|46|48|49|50|52|53|54|56|57|58|60|61|62)\.input_layernorm$
+        balance_layers: ['re:.*linear_attn.in_proj_qkv$', 're:.*linear_attn.in_proj_z$', 're:.*linear_attn.in_proj_b$',
+          're:.*linear_attn.in_proj_a$']
+        activation_hook_target: null
+      - smooth_layer: re:.*post_attention_layernorm$
+        balance_layers: ['re:.*gate_proj$', 're:.*up_proj$']
+        activation_hook_target: null
+      - smooth_layer: re:.*up_proj$
+        balance_layers: ['re:.*down_proj$']
+        activation_hook_target: null
+      offload_device: !!python/object/apply:torch.device [cpu]
+      duo_scaling: true
+      n_grid: 20
+    QuantizationModifier:
+      targets: [Linear]
+      ignore: ['re:.*lm_head', 're:model\.visual\..*', 're:.*\.linear_attn\.in_proj_a', 're:.*\.linear_attn\.in_proj_b',
+        're:mtp\..*']
+      scheme: W4A16_ASYM
+      bypass_divisibility_checks: false
+      requires_calibration_data: false

tokenizer.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:06b9509352d2af50381ab2247e083b80d32d5c0aba91c272ca9ff729b6a0e523
+size 19989325

tokenizer_config.json

@@ -0,0 +1,33 @@
+{
+  "add_prefix_space": false,
+  "audio_bos_token": "<|audio_start|>",
+  "audio_eos_token": "<|audio_end|>",
+  "audio_token": "<|audio_pad|>",
+  "backend": "tokenizers",
+  "bos_token": null,
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "<|im_end|>",
+  "errors": "replace",
+  "image_token": "<|image_pad|>",
+  "is_local": true,
+  "local_files_only": false,
+  "model_max_length": 262144,
+  "model_specific_special_tokens": {
+    "audio_bos_token": "<|audio_start|>",
+    "audio_eos_token": "<|audio_end|>",
+    "audio_token": "<|audio_pad|>",
+    "image_token": "<|image_pad|>",
+    "video_token": "<|video_pad|>",
+    "vision_bos_token": "<|vision_start|>",
+    "vision_eos_token": "<|vision_end|>"
+  },
+  "pad_token": "<|endoftext|>",
+  "pretokenize_regex": "(?i:'s|'t|'re|'ve|'m|'ll|'d)|[^\\r\\n\\p{L}\\p{N}]?[\\p{L}\\p{M}]+|\\p{N}| ?[^\\s\\p{L}\\p{M}\\p{N}]+[\\r\\n]*|\\s*[\\r\\n]+|\\s+(?!\\S)|\\s+",
+  "processor_class": "Qwen3VLProcessor",
+  "split_special_tokens": false,
+  "tokenizer_class": "Qwen2Tokenizer",
+  "unk_token": null,
+  "video_token": "<|video_pad|>",
+  "vision_bos_token": "<|vision_start|>",
+  "vision_eos_token": "<|vision_end|>"
+}

video_preprocessor_config.json

@@ -0,0 +1,21 @@
+{
+    "size": {
+        "longest_edge": 25165824,
+        "shortest_edge": 4096
+    },
+    "patch_size": 16,
+    "temporal_patch_size": 2,
+    "merge_size": 2,
+    "image_mean": [
+        0.5,
+        0.5,
+        0.5
+    ],
+    "image_std": [
+        0.5,
+        0.5,
+        0.5
+    ],
+    "processor_class": "Qwen3VLProcessor",
+    "video_processor_type": "Qwen3VLVideoProcessor"
+}