Text Generation
Transformers
ONNX
Safetensors
Arabic
English
qwen3
yemenjpt
osint
journalism
arabic
qwen
conversational
Instructions to use Yemen-JPT/Micro with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use Yemen-JPT/Micro with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("text-generation", model="Yemen-JPT/Micro") messages = [ {"role": "user", "content": "Who are you?"}, ] pipe(messages)# Load model directly from transformers import AutoTokenizer, AutoModelForCausalLM tokenizer = AutoTokenizer.from_pretrained("Yemen-JPT/Micro") model = AutoModelForCausalLM.from_pretrained("Yemen-JPT/Micro") messages = [ {"role": "user", "content": "Who are you?"}, ] inputs = tokenizer.apply_chat_template( messages, add_generation_prompt=True, tokenize=True, return_dict=True, return_tensors="pt", ).to(model.device) outputs = model.generate(**inputs, max_new_tokens=40) print(tokenizer.decode(outputs[0][inputs["input_ids"].shape[-1]:])) - Notebooks
- Google Colab
- Kaggle
- Local Apps Settings
- vLLM
How to use Yemen-JPT/Micro with vLLM:
Install from pip and serve model
# Install vLLM from pip: pip install vllm # Start the vLLM server: vllm serve "Yemen-JPT/Micro" # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:8000/v1/chat/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "Yemen-JPT/Micro", "messages": [ { "role": "user", "content": "What is the capital of France?" } ] }'Use Docker
docker model run hf.co/Yemen-JPT/Micro
- SGLang
How to use Yemen-JPT/Micro with SGLang:
Install from pip and serve model
# Install SGLang from pip: pip install sglang # Start the SGLang server: python3 -m sglang.launch_server \ --model-path "Yemen-JPT/Micro" \ --host 0.0.0.0 \ --port 30000 # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:30000/v1/chat/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "Yemen-JPT/Micro", "messages": [ { "role": "user", "content": "What is the capital of France?" } ] }'Use Docker images
docker run --gpus all \ --shm-size 32g \ -p 30000:30000 \ -v ~/.cache/huggingface:/root/.cache/huggingface \ --env "HF_TOKEN=<secret>" \ --ipc=host \ lmsysorg/sglang:latest \ python3 -m sglang.launch_server \ --model-path "Yemen-JPT/Micro" \ --host 0.0.0.0 \ --port 30000 # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:30000/v1/chat/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "Yemen-JPT/Micro", "messages": [ { "role": "user", "content": "What is the capital of France?" } ] }' - Docker Model Runner
How to use Yemen-JPT/Micro with Docker Model Runner:
docker model run hf.co/Yemen-JPT/Micro
Ameen commited on
Upload OSINT-Micro - Lightweight ONNX model for microservices (5.7GB)
Browse files- .gitattributes +3 -0
- README.md +52 -0
- added_tokens.json +29 -0
- chat_template.jinja +97 -0
- config.json +81 -0
- generation_config.json +6 -0
- merges.txt +0 -0
- onnx/model_q4f16.onnx +3 -0
- onnx/model_q4f16.onnx_data +3 -0
- onnx/model_q4f16.onnx_data_1 +3 -0
- special_tokens_map.json +16 -0
- tokenizer.json +3 -0
- tokenizer_config.json +16 -0
- vocab.json +0 -0
.gitattributes
CHANGED
|
@@ -33,3 +33,6 @@ 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 |
+
onnx/model_q4f16.onnx_data filter=lfs diff=lfs merge=lfs -text
|
| 37 |
+
onnx/model_q4f16.onnx_data_1 filter=lfs diff=lfs merge=lfs -text
|
| 38 |
+
tokenizer.json filter=lfs diff=lfs merge=lfs -text
|
README.md
ADDED
|
@@ -0,0 +1,52 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
license: apache-2.0
|
| 3 |
+
base_model: Qwen/Qwen3-4B
|
| 4 |
+
tags:
|
| 5 |
+
- qwen3
|
| 6 |
+
- investigative-journalism
|
| 7 |
+
- osint
|
| 8 |
+
- onnx
|
| 9 |
+
- transformers.js
|
| 10 |
+
- webgpu
|
| 11 |
+
- conversational
|
| 12 |
+
---
|
| 13 |
+
|
| 14 |
+
# qwen3-4b-journalist-ONNX
|
| 15 |
+
|
| 16 |
+
An investigative journalism model fine-tuned on [Qwen/Qwen3-4B](https://huggingface.co/Qwen/Qwen3-4B), converted to ONNX for in-browser inference via [Transformers.js](https://github.com/huggingface/transformers.js) and WebGPU.
|
| 17 |
+
|
| 18 |
+
Built by [Buried Signals](https://buriedsignals.com) for local/browser OSINT assistant workflows.
|
| 19 |
+
|
| 20 |
+
## Usage
|
| 21 |
+
|
| 22 |
+
```javascript
|
| 23 |
+
import { AutoTokenizer, AutoModelForCausalLM } from "@huggingface/transformers";
|
| 24 |
+
|
| 25 |
+
const tokenizer = await AutoTokenizer.from_pretrained("tomvaillant/qwen3-4b-journalist-ONNX");
|
| 26 |
+
const model = await AutoModelForCausalLM.from_pretrained("tomvaillant/qwen3-4b-journalist-ONNX", {
|
| 27 |
+
dtype: "q4f16",
|
| 28 |
+
device: "webgpu",
|
| 29 |
+
});
|
| 30 |
+
```
|
| 31 |
+
|
| 32 |
+
## Training
|
| 33 |
+
|
| 34 |
+
- **Method:** QLoRA with [Unsloth](https://github.com/unslothai/unsloth) / TRL SFT
|
| 35 |
+
- **Data:** 700 instruction/response pairs across investigative journalism and OSINT categories
|
| 36 |
+
- **Task:** browser-sized assistant for OSINT tool selection, investigation planning, verification, and source handling
|
| 37 |
+
|
| 38 |
+
## ONNX Conversion
|
| 39 |
+
|
| 40 |
+
- **Quantization:** q4f16
|
| 41 |
+
- **Runtime:** Transformers.js via WebGPU
|
| 42 |
+
- **Files:** `onnx/model_q4f16.onnx`, external data files, tokenizer, and chat template
|
| 43 |
+
|
| 44 |
+
## Sources And Attribution
|
| 45 |
+
|
| 46 |
+
Training data was synthesized from the same investigative journalism corpus as the Gemma variants. It includes [tomvaillant/osint-tool-database](https://huggingface.co/datasets/tomvaillant/osint-tool-database), Indicator Media briefing tools, Buried Signals investigative skills, GIJN manuals, Bellingcat guides, Verification Handbook 3, SPJ Code of Ethics, RCFP resources, and public investigative manuals from UNESCO, Al Jazeera Media Institute, CiFAR, CIPE, and EJF/TEMPO Institute.
|
| 47 |
+
|
| 48 |
+
Full attribution is maintained in the training repository as `SOURCES.md`.
|
| 49 |
+
|
| 50 |
+
## Intended Use
|
| 51 |
+
|
| 52 |
+
Use for local/browser investigative assistance. Generated findings, URLs, and legal conclusions require independent verification before reporting or operational use.
|
added_tokens.json
ADDED
|
@@ -0,0 +1,29 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"</think>": 151668,
|
| 3 |
+
"</tool_call>": 151658,
|
| 4 |
+
"</tool_response>": 151666,
|
| 5 |
+
"<think>": 151667,
|
| 6 |
+
"<tool_call>": 151657,
|
| 7 |
+
"<tool_response>": 151665,
|
| 8 |
+
"<|PAD_TOKEN|>": 151669,
|
| 9 |
+
"<|box_end|>": 151649,
|
| 10 |
+
"<|box_start|>": 151648,
|
| 11 |
+
"<|endoftext|>": 151643,
|
| 12 |
+
"<|file_sep|>": 151664,
|
| 13 |
+
"<|fim_middle|>": 151660,
|
| 14 |
+
"<|fim_pad|>": 151662,
|
| 15 |
+
"<|fim_prefix|>": 151659,
|
| 16 |
+
"<|fim_suffix|>": 151661,
|
| 17 |
+
"<|im_end|>": 151645,
|
| 18 |
+
"<|im_start|>": 151644,
|
| 19 |
+
"<|image_pad|>": 151655,
|
| 20 |
+
"<|object_ref_end|>": 151647,
|
| 21 |
+
"<|object_ref_start|>": 151646,
|
| 22 |
+
"<|quad_end|>": 151651,
|
| 23 |
+
"<|quad_start|>": 151650,
|
| 24 |
+
"<|repo_name|>": 151663,
|
| 25 |
+
"<|video_pad|>": 151656,
|
| 26 |
+
"<|vision_end|>": 151653,
|
| 27 |
+
"<|vision_pad|>": 151654,
|
| 28 |
+
"<|vision_start|>": 151652
|
| 29 |
+
}
|
chat_template.jinja
ADDED
|
@@ -0,0 +1,97 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{%- if tools %}
|
| 2 |
+
{{- '<|im_start|>system\n' }}
|
| 3 |
+
{%- if messages[0].role == 'system' %}
|
| 4 |
+
{{- messages[0].content + '\n\n' }}
|
| 5 |
+
{%- endif %}
|
| 6 |
+
{{- "# 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>" }}
|
| 7 |
+
{%- for tool in tools %}
|
| 8 |
+
{{- "\n" }}
|
| 9 |
+
{{- tool | tojson }}
|
| 10 |
+
{%- endfor %}
|
| 11 |
+
{{- "\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" }}
|
| 12 |
+
{%- else %}
|
| 13 |
+
{%- if messages[0].role == 'system' %}
|
| 14 |
+
{{- '<|im_start|>system\n' + messages[0].content + '<|im_end|>\n' }}
|
| 15 |
+
{%- endif %}
|
| 16 |
+
{%- endif %}
|
| 17 |
+
{%- set ns = namespace(multi_step_tool=true, last_query_index=messages|length - 1) %}
|
| 18 |
+
{%- for forward_message in messages %}
|
| 19 |
+
{%- set index = (messages|length - 1) - loop.index0 %}
|
| 20 |
+
{%- set message = messages[index] %}
|
| 21 |
+
{%- set tool_start = '<tool_response>' %}
|
| 22 |
+
{%- set tool_start_length = tool_start|length %}
|
| 23 |
+
{%- set start_of_message = message.content[:tool_start_length] %}
|
| 24 |
+
{%- set tool_end = '</tool_response>' %}
|
| 25 |
+
{%- set tool_end_length = tool_end|length %}
|
| 26 |
+
{%- set start_pos = (message.content|length) - tool_end_length %}
|
| 27 |
+
{%- if start_pos < 0 %}
|
| 28 |
+
{%- set start_pos = 0 %}
|
| 29 |
+
{%- endif %}
|
| 30 |
+
{%- set end_of_message = message.content[start_pos:] %}
|
| 31 |
+
{%- if ns.multi_step_tool and message.role == "user" and not(start_of_message == tool_start and end_of_message == tool_end) %}
|
| 32 |
+
{%- set ns.multi_step_tool = false %}
|
| 33 |
+
{%- set ns.last_query_index = index %}
|
| 34 |
+
{%- endif %}
|
| 35 |
+
{%- endfor %}
|
| 36 |
+
{%- for message in messages %}
|
| 37 |
+
{%- if (message.role == "user") or (message.role == "system" and not loop.first) %}
|
| 38 |
+
{{- '<|im_start|>' + message.role + '\n' + message.content + '<|im_end|>' + '\n' }}
|
| 39 |
+
{%- elif message.role == "assistant" %}
|
| 40 |
+
{%- set content = message.content %}
|
| 41 |
+
{%- set reasoning_content = '' %}
|
| 42 |
+
{%- if message.reasoning_content is defined and message.reasoning_content is not none %}
|
| 43 |
+
{%- set reasoning_content = message.reasoning_content %}
|
| 44 |
+
{%- else %}
|
| 45 |
+
{%- if '</think>' in message.content %}
|
| 46 |
+
{%- set content = (message.content.split('</think>')|last).lstrip('\n') %}
|
| 47 |
+
{%- set reasoning_content = (message.content.split('</think>')|first).rstrip('\n') %}
|
| 48 |
+
{%- set reasoning_content = (reasoning_content.split('<think>')|last).lstrip('\n') %}
|
| 49 |
+
{%- endif %}
|
| 50 |
+
{%- endif %}
|
| 51 |
+
{%- if loop.index0 > ns.last_query_index %}
|
| 52 |
+
{%- if loop.last or (not loop.last and reasoning_content) %}
|
| 53 |
+
{{- '<|im_start|>' + message.role + '\n<think>\n' + reasoning_content.strip('\n') + '\n</think>\n\n' + content.lstrip('\n') }}
|
| 54 |
+
{%- else %}
|
| 55 |
+
{{- '<|im_start|>' + message.role + '\n' + content }}
|
| 56 |
+
{%- endif %}
|
| 57 |
+
{%- else %}
|
| 58 |
+
{{- '<|im_start|>' + message.role + '\n' + content }}
|
| 59 |
+
{%- endif %}
|
| 60 |
+
{%- if message.tool_calls %}
|
| 61 |
+
{%- for tool_call in message.tool_calls %}
|
| 62 |
+
{%- if (loop.first and content) or (not loop.first) %}
|
| 63 |
+
{{- '\n' }}
|
| 64 |
+
{%- endif %}
|
| 65 |
+
{%- if tool_call.function %}
|
| 66 |
+
{%- set tool_call = tool_call.function %}
|
| 67 |
+
{%- endif %}
|
| 68 |
+
{{- '<tool_call>\n{"name": "' }}
|
| 69 |
+
{{- tool_call.name }}
|
| 70 |
+
{{- '", "arguments": ' }}
|
| 71 |
+
{%- if tool_call.arguments is string %}
|
| 72 |
+
{{- tool_call.arguments }}
|
| 73 |
+
{%- else %}
|
| 74 |
+
{{- tool_call.arguments | tojson }}
|
| 75 |
+
{%- endif %}
|
| 76 |
+
{{- '}\n</tool_call>' }}
|
| 77 |
+
{%- endfor %}
|
| 78 |
+
{%- endif %}
|
| 79 |
+
{{- '<|im_end|>\n' }}
|
| 80 |
+
{%- elif message.role == "tool" %}
|
| 81 |
+
{%- if loop.first or (messages[loop.index0 - 1].role != "tool") %}
|
| 82 |
+
{{- '<|im_start|>user' }}
|
| 83 |
+
{%- endif %}
|
| 84 |
+
{{- '\n<tool_response>\n' }}
|
| 85 |
+
{{- message.content }}
|
| 86 |
+
{{- '\n</tool_response>' }}
|
| 87 |
+
{%- if loop.last or (messages[loop.index0 + 1].role != "tool") %}
|
| 88 |
+
{{- '<|im_end|>\n' }}
|
| 89 |
+
{%- endif %}
|
| 90 |
+
{%- endif %}
|
| 91 |
+
{%- endfor %}
|
| 92 |
+
{%- if add_generation_prompt %}
|
| 93 |
+
{{- '<|im_start|>assistant\n' }}
|
| 94 |
+
{%- if enable_thinking is defined and enable_thinking is false %}
|
| 95 |
+
{{- '<think>\n\n</think>\n\n' }}
|
| 96 |
+
{%- endif %}
|
| 97 |
+
{%- endif %}
|
config.json
ADDED
|
@@ -0,0 +1,81 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"architectures": [
|
| 3 |
+
"Qwen3ForCausalLM"
|
| 4 |
+
],
|
| 5 |
+
"attention_bias": false,
|
| 6 |
+
"attention_dropout": 0.0,
|
| 7 |
+
"bos_token_id": null,
|
| 8 |
+
"torch_dtype": "bfloat16",
|
| 9 |
+
"eos_token_id": 151645,
|
| 10 |
+
"head_dim": 128,
|
| 11 |
+
"hidden_act": "silu",
|
| 12 |
+
"hidden_size": 2560,
|
| 13 |
+
"initializer_range": 0.02,
|
| 14 |
+
"intermediate_size": 9728,
|
| 15 |
+
"layer_types": [
|
| 16 |
+
"full_attention",
|
| 17 |
+
"full_attention",
|
| 18 |
+
"full_attention",
|
| 19 |
+
"full_attention",
|
| 20 |
+
"full_attention",
|
| 21 |
+
"full_attention",
|
| 22 |
+
"full_attention",
|
| 23 |
+
"full_attention",
|
| 24 |
+
"full_attention",
|
| 25 |
+
"full_attention",
|
| 26 |
+
"full_attention",
|
| 27 |
+
"full_attention",
|
| 28 |
+
"full_attention",
|
| 29 |
+
"full_attention",
|
| 30 |
+
"full_attention",
|
| 31 |
+
"full_attention",
|
| 32 |
+
"full_attention",
|
| 33 |
+
"full_attention",
|
| 34 |
+
"full_attention",
|
| 35 |
+
"full_attention",
|
| 36 |
+
"full_attention",
|
| 37 |
+
"full_attention",
|
| 38 |
+
"full_attention",
|
| 39 |
+
"full_attention",
|
| 40 |
+
"full_attention",
|
| 41 |
+
"full_attention",
|
| 42 |
+
"full_attention",
|
| 43 |
+
"full_attention",
|
| 44 |
+
"full_attention",
|
| 45 |
+
"full_attention",
|
| 46 |
+
"full_attention",
|
| 47 |
+
"full_attention",
|
| 48 |
+
"full_attention",
|
| 49 |
+
"full_attention",
|
| 50 |
+
"full_attention",
|
| 51 |
+
"full_attention"
|
| 52 |
+
],
|
| 53 |
+
"max_position_embeddings": 40960,
|
| 54 |
+
"max_window_layers": 36,
|
| 55 |
+
"model_type": "qwen3",
|
| 56 |
+
"num_attention_heads": 32,
|
| 57 |
+
"num_hidden_layers": 36,
|
| 58 |
+
"num_key_value_heads": 8,
|
| 59 |
+
"pad_token_id": 151669,
|
| 60 |
+
"rms_norm_eps": 1e-06,
|
| 61 |
+
"rope_parameters": {
|
| 62 |
+
"rope_theta": 1000000,
|
| 63 |
+
"rope_type": "default"
|
| 64 |
+
},
|
| 65 |
+
"sliding_window": null,
|
| 66 |
+
"tie_word_embeddings": true,
|
| 67 |
+
"unsloth_fixed": true,
|
| 68 |
+
"unsloth_version": "2026.4.4",
|
| 69 |
+
"use_cache": true,
|
| 70 |
+
"use_sliding_window": false,
|
| 71 |
+
"vocab_size": 151936,
|
| 72 |
+
"transformers.js_config": {
|
| 73 |
+
"dtype": "q4f16",
|
| 74 |
+
"kv_cache_dtype": {
|
| 75 |
+
"q4f16": "float16"
|
| 76 |
+
},
|
| 77 |
+
"use_external_data_format": {
|
| 78 |
+
"model_q4f16.onnx": 2
|
| 79 |
+
}
|
| 80 |
+
}
|
| 81 |
+
}
|
generation_config.json
ADDED
|
@@ -0,0 +1,6 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_from_model_config": true,
|
| 3 |
+
"eos_token_id": 151645,
|
| 4 |
+
"pad_token_id": 151669,
|
| 5 |
+
"transformers_version": "4.57.6"
|
| 6 |
+
}
|
merges.txt
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
onnx/model_q4f16.onnx
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:bebdbe2f5d4fb822cb090ccc55f4470e90889532867372724d3684cb4d767282
|
| 3 |
+
size 2583451
|
onnx/model_q4f16.onnx_data
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:f9dcb7bae0caa725a641e90602a4dfe72abc1769f9d5d2560007540a8b03c9bf
|
| 3 |
+
size 1899530240
|
onnx/model_q4f16.onnx_data_1
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:52b7043779ea76cca528a047130c14319ae3bfcf4f6dbb32930347d32e936d84
|
| 3 |
+
size 1141302272
|
special_tokens_map.json
ADDED
|
@@ -0,0 +1,16 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"eos_token": {
|
| 3 |
+
"content": "<|im_end|>",
|
| 4 |
+
"lstrip": false,
|
| 5 |
+
"normalized": false,
|
| 6 |
+
"rstrip": false,
|
| 7 |
+
"single_word": false
|
| 8 |
+
},
|
| 9 |
+
"pad_token": {
|
| 10 |
+
"content": "<|PAD_TOKEN|>",
|
| 11 |
+
"lstrip": false,
|
| 12 |
+
"normalized": false,
|
| 13 |
+
"rstrip": false,
|
| 14 |
+
"single_word": false
|
| 15 |
+
}
|
| 16 |
+
}
|
tokenizer.json
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:d7430e9138b76e93fb6f93462394d236b411111aef53cb421ba97d2691040cca
|
| 3 |
+
size 11423114
|
tokenizer_config.json
ADDED
|
@@ -0,0 +1,16 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"add_prefix_space": false,
|
| 3 |
+
"backend": "tokenizers",
|
| 4 |
+
"bos_token": null,
|
| 5 |
+
"clean_up_tokenization_spaces": false,
|
| 6 |
+
"eos_token": "<|im_end|>",
|
| 7 |
+
"errors": "replace",
|
| 8 |
+
"is_local": false,
|
| 9 |
+
"model_max_length": 40960,
|
| 10 |
+
"pad_token": "<|PAD_TOKEN|>",
|
| 11 |
+
"padding_side": "right",
|
| 12 |
+
"split_special_tokens": false,
|
| 13 |
+
"tokenizer_class": "Qwen2Tokenizer",
|
| 14 |
+
"unk_token": null,
|
| 15 |
+
"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 forward_message in messages %}\n {%- set index = (messages|length - 1) - loop.index0 %}\n {%- set message = messages[index] %}\n {%- set tool_start = '<tool_response>' %}\n {%- set tool_start_length = tool_start|length %}\n {%- set start_of_message = message.content[:tool_start_length] %}\n {%- set tool_end = '</tool_response>' %}\n {%- set tool_end_length = tool_end|length %}\n {%- set start_pos = (message.content|length) - tool_end_length %}\n {%- if start_pos < 0 %}\n {%- set start_pos = 0 %}\n {%- endif %}\n {%- set end_of_message = message.content[start_pos:] %}\n {%- if ns.multi_step_tool and message.role == \"user\" and not(start_of_message == tool_start and end_of_message == tool_end) %}\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.role == \"user\") or (message.role == \"system\" and not loop.first) %}\n {{- '<|im_start|>' + message.role + '\\n' + message.content + '<|im_end|>' + '\\n' }}\n {%- elif message.role == \"assistant\" %}\n {%- set content = message.content %}\n {%- set reasoning_content = '' %}\n {%- if message.reasoning_content is defined and message.reasoning_content is not none %}\n {%- set reasoning_content = message.reasoning_content %}\n {%- else %}\n {%- if '</think>' in message.content %}\n {%- set content = (message.content.split('</think>')|last).lstrip('\\n') %}\n {%- set reasoning_content = (message.content.split('</think>')|first).rstrip('\\n') %}\n {%- set reasoning_content = (reasoning_content.split('<think>')|last).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 {{- message.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 %}"
|
| 16 |
+
}
|
vocab.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|