Upload folder using huggingface_hub
Browse files- .gitattributes +2 -0
- README.md +25 -0
- config.json +16 -0
- config_cpu.json +9 -0
- draft_config_cpu.json +9 -0
- export_args.json +42 -0
- llm.mnn +3 -0
- llm.mnn.weight +3 -0
- llm_config.json +35 -0
- tokenizer.txt +0 -0
.gitattributes
CHANGED
|
@@ -33,3 +33,5 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
|
|
| 33 |
*.zip filter=lfs diff=lfs merge=lfs -text
|
| 34 |
*.zst filter=lfs diff=lfs merge=lfs -text
|
| 35 |
*tfevents* filter=lfs diff=lfs merge=lfs -text
|
|
|
|
|
|
|
|
|
| 33 |
*.zip filter=lfs diff=lfs merge=lfs -text
|
| 34 |
*.zst filter=lfs diff=lfs merge=lfs -text
|
| 35 |
*tfevents* filter=lfs diff=lfs merge=lfs -text
|
| 36 |
+
llm.mnn filter=lfs diff=lfs merge=lfs -text
|
| 37 |
+
llm.mnn.weight filter=lfs diff=lfs merge=lfs -text
|
README.md
ADDED
|
@@ -0,0 +1,25 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
# TokForge Acceleration Pack — Qwen3.5 Draft Model
|
| 2 |
+
|
| 3 |
+
**Qwen3.5-0.8B Abliterated (MNN)** draft model for speculative decoding in [TokForge](https://tokforge.ai).
|
| 4 |
+
|
| 5 |
+
Based on [huihui-ai/Huihui-Qwen3.5-0.8B-abliterated](https://huggingface.co/huihui-ai/Huihui-Qwen3.5-0.8B-abliterated).
|
| 6 |
+
|
| 7 |
+
## Conversion Details
|
| 8 |
+
- **Quantization**: 4-bit HQQ (quant_block=64)
|
| 9 |
+
- **Format**: MNN (text-only, visual encoder removed)
|
| 10 |
+
- **Model type**: qwen3_5 (LinearAttention architecture)
|
| 11 |
+
- **Weight size**: ~449MB
|
| 12 |
+
|
| 13 |
+
## Compatible Target Models
|
| 14 |
+
- Qwen3.5-4B (MNN)
|
| 15 |
+
- Qwen3.5-9B (MNN)
|
| 16 |
+
- Qwen3.5-27B (MNN)
|
| 17 |
+
- Any Qwen3.5-based abliterated/uncensored model
|
| 18 |
+
|
| 19 |
+
## Note
|
| 20 |
+
Qwen3.5 models use LinearAttention which runs on CPU. Spec decode with CPU draft + CPU target has different economics than GPU targets — testing is needed to verify benefit on each SoC.
|
| 21 |
+
|
| 22 |
+
## Attribution
|
| 23 |
+
- Original model: [Qwen/Qwen3.5-0.8B](https://huggingface.co/Qwen/Qwen3.5-0.8B)
|
| 24 |
+
- Abliteration: [huihui-ai](https://huggingface.co/huihui-ai)
|
| 25 |
+
- MNN conversion: [taobao-mnn](https://huggingface.co/taobao-mnn) pipeline / darkmaniac7
|
config.json
ADDED
|
@@ -0,0 +1,16 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"llm_model": "llm.mnn",
|
| 3 |
+
"llm_weight": "llm.mnn.weight",
|
| 4 |
+
"backend_type": "cpu",
|
| 5 |
+
"thread_num": 4,
|
| 6 |
+
"precision": "low",
|
| 7 |
+
"memory": "low",
|
| 8 |
+
"sampler_type": "penalty",
|
| 9 |
+
"penalty": 1.1,
|
| 10 |
+
"mllm": {
|
| 11 |
+
"backend_type": "cpu",
|
| 12 |
+
"thread_num": 4,
|
| 13 |
+
"precision": "normal",
|
| 14 |
+
"memory": "low"
|
| 15 |
+
}
|
| 16 |
+
}
|
config_cpu.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"llm_model": "llm.mnn",
|
| 3 |
+
"llm_weight": "llm.mnn.weight",
|
| 4 |
+
"backend_type": "cpu",
|
| 5 |
+
"thread_num": 1,
|
| 6 |
+
"precision": "low",
|
| 7 |
+
"power": "high",
|
| 8 |
+
"sampler_type": "greedy"
|
| 9 |
+
}
|
draft_config_cpu.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"llm_model": "llm.mnn",
|
| 3 |
+
"llm_weight": "llm.mnn.weight",
|
| 4 |
+
"backend_type": "cpu",
|
| 5 |
+
"thread_num": 1,
|
| 6 |
+
"precision": "low",
|
| 7 |
+
"power": "high",
|
| 8 |
+
"sampler_type": "greedy"
|
| 9 |
+
}
|
export_args.json
ADDED
|
@@ -0,0 +1,42 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"path": "/tmp/qwen35-abliterated-hf",
|
| 3 |
+
"type": null,
|
| 4 |
+
"tokenizer_path": "/tmp/qwen35-abliterated-hf",
|
| 5 |
+
"eagle_path": null,
|
| 6 |
+
"lora_path": null,
|
| 7 |
+
"gptq_path": null,
|
| 8 |
+
"dst_path": "/tmp/qwen35-draft",
|
| 9 |
+
"verbose": false,
|
| 10 |
+
"test": null,
|
| 11 |
+
"export": "mnn",
|
| 12 |
+
"onnx_slim": false,
|
| 13 |
+
"quant_bit": 4,
|
| 14 |
+
"quant_block": 64,
|
| 15 |
+
"visual_quant_bit": null,
|
| 16 |
+
"visual_quant_block": null,
|
| 17 |
+
"lm_quant_bit": 4,
|
| 18 |
+
"lm_quant_block": 64,
|
| 19 |
+
"mnnconvert": "/root/MNN-tokforge/build-host/MNNConvert",
|
| 20 |
+
"ppl": false,
|
| 21 |
+
"awq": false,
|
| 22 |
+
"hqq": true,
|
| 23 |
+
"omni": false,
|
| 24 |
+
"transformer_fuse": false,
|
| 25 |
+
"group_conv_native": false,
|
| 26 |
+
"smooth": false,
|
| 27 |
+
"sym": false,
|
| 28 |
+
"visual_sym": false,
|
| 29 |
+
"seperate_embed": false,
|
| 30 |
+
"lora_split": false,
|
| 31 |
+
"calib_data": null,
|
| 32 |
+
"act_bit": 16,
|
| 33 |
+
"embed_bit": 16,
|
| 34 |
+
"act_sym": false,
|
| 35 |
+
"quant_config": null,
|
| 36 |
+
"generate_for_npu": false,
|
| 37 |
+
"skip_weight": false,
|
| 38 |
+
"omni_epochs": 20,
|
| 39 |
+
"omni_lr": 0.005,
|
| 40 |
+
"omni_wd": 0.0001,
|
| 41 |
+
"tie_word_embeddings": true
|
| 42 |
+
}
|
llm.mnn
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:8cc883ae68f564a0ca34c8c3e0238abfdac37c67b9c672b0f08d1690b32e7716
|
| 3 |
+
size 2148136
|
llm.mnn.weight
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:4ce1b103ee6972475ee857d50f5bfbe964d82ec894a42107815f584eb96d61a1
|
| 3 |
+
size 470382614
|
llm_config.json
ADDED
|
@@ -0,0 +1,35 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"model_type": "qwen3_5",
|
| 3 |
+
"hidden_size": 1024,
|
| 4 |
+
"attention_mask": "float",
|
| 5 |
+
"attention_type": "full",
|
| 6 |
+
"is_mrope": true,
|
| 7 |
+
"is_visual": true,
|
| 8 |
+
"image_mean": [
|
| 9 |
+
127.5,
|
| 10 |
+
127.5,
|
| 11 |
+
127.5
|
| 12 |
+
],
|
| 13 |
+
"image_norm": [
|
| 14 |
+
0.00784313725490196,
|
| 15 |
+
0.00784313725490196,
|
| 16 |
+
0.00784313725490196
|
| 17 |
+
],
|
| 18 |
+
"image_size": 420,
|
| 19 |
+
"vision_start": 248053,
|
| 20 |
+
"vision_end": 248054,
|
| 21 |
+
"image_pad": 248056,
|
| 22 |
+
"num_grid_per_side": 48,
|
| 23 |
+
"has_deepstack": true,
|
| 24 |
+
"jinja": {
|
| 25 |
+
"chat_template": "{%- set image_count = namespace(value=0) %}\n{%- set video_count = namespace(value=0) %}\n{%- macro render_content(content, do_vision_count, is_system_content=false) %}\n {%- if content is string %}\n {{- content }}\n {%- elif content is iterable and content is not mapping %}\n {%- for item in content %}\n {%- if 'image' in item or 'image_url' in item or item.type == 'image' %}\n {%- if is_system_content %}\n {{- raise_exception('System message cannot contain images.') }}\n {%- endif %}\n {%- if do_vision_count %}\n {%- set image_count.value = image_count.value + 1 %}\n {%- endif %}\n {%- if add_vision_id %}\n {{- 'Picture ' ~ image_count.value ~ ': ' }}\n {%- endif %}\n {{- '<|vision_start|><|image_pad|><|vision_end|>' }}\n {%- elif 'video' in item or item.type == 'video' %}\n {%- if is_system_content %}\n {{- raise_exception('System message cannot contain videos.') }}\n {%- endif %}\n {%- if do_vision_count %}\n {%- set video_count.value = video_count.value + 1 %}\n {%- endif %}\n {%- if add_vision_id %}\n {{- 'Video ' ~ video_count.value ~ ': ' }}\n {%- endif %}\n {{- '<|vision_start|><|video_pad|><|vision_end|>' }}\n {%- elif 'text' in item %}\n {{- item.text }}\n {%- else %}\n {{- raise_exception('Unexpected item type in content.') }}\n {%- endif %}\n {%- endfor %}\n {%- elif content is none or content is undefined %}\n {{- '' }}\n {%- else %}\n {{- raise_exception('Unexpected content type.') }}\n {%- endif %}\n{%- endmacro %}\n{%- if not messages %}\n {{- raise_exception('No messages provided.') }}\n{%- endif %}\n{%- if tools and tools is iterable and tools is not mapping %}\n {{- '<|im_start|>system\\n' }}\n {{- \"# Tools\\n\\nYou have access to the following functions:\\n\\n<tools>\" }}\n {%- for tool in tools %}\n {{- \"\\n\" }}\n {{- tool | tojson }}\n {%- endfor %}\n {{- \"\\n</tools>\" }}\n {{- '\\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>' }}\n {%- if messages[0].role == 'system' %}\n {%- set content = render_content(messages[0].content, false, true)|trim %}\n {%- if content %}\n {{- '\\n\\n' + content }}\n {%- endif %}\n {%- endif %}\n {{- '<|im_end|>\\n' }}\n{%- else %}\n {%- if messages[0].role == 'system' %}\n {%- set content = render_content(messages[0].content, false, true)|trim %}\n {{- '<|im_start|>system\\n' + 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\" %}\n {%- set content = render_content(message.content, false)|trim %}\n {%- if not(content.startswith('<tool_response>') and content.endswith('</tool_response>')) %}\n {%- set ns.multi_step_tool = false %}\n {%- set ns.last_query_index = index %}\n {%- endif %}\n {%- endif %}\n{%- endfor %}\n{%- if ns.multi_step_tool %}\n {{- raise_exception('No user query found in messages.') }}\n{%- endif %}\n{%- for message in messages %}\n {%- set content = render_content(message.content, true)|trim %}\n {%- if message.role == \"system\" %}\n {%- if not loop.first %}\n {{- raise_exception('System message must be at the beginning.') }}\n {%- endif %}\n {%- elif message.role == \"user\" %}\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 {%- set reasoning_content = reasoning_content|trim %}\n {%- if loop.index0 > ns.last_query_index %}\n {{- '<|im_start|>' + message.role + '\\n<think>\\n' + reasoning_content + '\\n</think>\\n\\n' + content }}\n {%- else %}\n {{- '<|im_start|>' + message.role + '\\n' + content }}\n {%- endif %}\n {%- if message.tool_calls and message.tool_calls is iterable and message.tool_calls is not mapping %}\n {%- for tool_call in message.tool_calls %}\n {%- if tool_call.function is defined %}\n {%- set tool_call = tool_call.function %}\n {%- endif %}\n {%- if loop.first %}\n {%- if content|trim %}\n {{- '\\n\\n<tool_call>\\n<function=' + tool_call.name + '>\\n' }}\n {%- else %}\n {{- '<tool_call>\\n<function=' + tool_call.name + '>\\n' }}\n {%- endif %}\n {%- else %}\n {{- '\\n<tool_call>\\n<function=' + tool_call.name + '>\\n' }}\n {%- endif %}\n {%- if tool_call.arguments is defined %}\n {%- for args_name, args_value in tool_call.arguments|items %}\n {{- '<parameter=' + args_name + '>\\n' }}\n {%- set args_value = args_value | tojson | safe if args_value is mapping or (args_value is sequence and args_value is not string) else args_value | string %}\n {{- args_value }}\n {{- '\\n</parameter>\\n' }}\n {%- endfor %}\n {%- endif %}\n {{- '</function>\\n</tool_call>' }}\n {%- endfor %}\n {%- endif %}\n {{- '<|im_end|>\\n' }}\n {%- elif message.role == \"tool\" %}\n {%- if loop.previtem and loop.previtem.role != \"tool\" %}\n {{- '<|im_start|>user' }}\n {%- endif %}\n {{- '\\n<tool_response>\\n' }}\n {{- content }}\n {{- '\\n</tool_response>' }}\n {%- if not loop.last and loop.nextitem.role != \"tool\" %}\n {{- '<|im_end|>\\n' }}\n {%- elif loop.last %}\n {{- '<|im_end|>\\n' }}\n {%- endif %}\n {%- else %}\n {{- raise_exception('Unexpected message role.') }}\n {%- endif %}\n{%- endfor %}\n{%- if add_generation_prompt %}\n {{- '<|im_start|>assistant\\n' }}\n {%- if enable_thinking is defined and enable_thinking is true %}\n {{- '<think>\\n' }}\n {%- else %}\n {{- '<think>\\n\\n</think>\\n\\n' }}\n {%- endif %}\n{%- endif %}",
|
| 26 |
+
"eos": "<|im_end|>"
|
| 27 |
+
},
|
| 28 |
+
"tie_embeddings": [
|
| 29 |
+
311457814,
|
| 30 |
+
438597654,
|
| 31 |
+
31784960,
|
| 32 |
+
4,
|
| 33 |
+
64
|
| 34 |
+
]
|
| 35 |
+
}
|
tokenizer.txt
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|