Simon-Liu/github-mcp-call-reasoning-1k
Viewer • Updated • 1k • 55
How to use Simon-Liu/gemma-4-e4b-github-mcp-sft-it with Transformers:
# Use a pipeline as a high-level helper
from transformers import pipeline
pipe = pipeline("text-generation", model="Simon-Liu/gemma-4-e4b-github-mcp-sft-it")
messages = [
{
"role": "user",
"content": [
{"type": "image", "url": "https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/p-blog/candy.JPG"},
{"type": "text", "text": "What animal is on the candy?"}
]
},
]
pipe(text=messages) # Load model directly
from transformers import AutoProcessor, AutoModelForMultimodalLM
processor = AutoProcessor.from_pretrained("Simon-Liu/gemma-4-e4b-github-mcp-sft-it")
model = AutoModelForMultimodalLM.from_pretrained("Simon-Liu/gemma-4-e4b-github-mcp-sft-it")
messages = [
{
"role": "user",
"content": [
{"type": "image", "url": "https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/p-blog/candy.JPG"},
{"type": "text", "text": "What animal is on the candy?"}
]
},
]
inputs = processor.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(processor.decode(outputs[0][inputs["input_ids"].shape[-1]:]))How to use Simon-Liu/gemma-4-e4b-github-mcp-sft-it with vLLM:
# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "Simon-Liu/gemma-4-e4b-github-mcp-sft-it"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
-H "Content-Type: application/json" \
--data '{
"model": "Simon-Liu/gemma-4-e4b-github-mcp-sft-it",
"messages": [
{
"role": "user",
"content": "What is the capital of France?"
}
]
}'docker model run hf.co/Simon-Liu/gemma-4-e4b-github-mcp-sft-it
How to use Simon-Liu/gemma-4-e4b-github-mcp-sft-it with SGLang:
# Install SGLang from pip:
pip install sglang
# Start the SGLang server:
python3 -m sglang.launch_server \
--model-path "Simon-Liu/gemma-4-e4b-github-mcp-sft-it" \
--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": "Simon-Liu/gemma-4-e4b-github-mcp-sft-it",
"messages": [
{
"role": "user",
"content": "What is the capital of France?"
}
]
}'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 "Simon-Liu/gemma-4-e4b-github-mcp-sft-it" \
--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": "Simon-Liu/gemma-4-e4b-github-mcp-sft-it",
"messages": [
{
"role": "user",
"content": "What is the capital of France?"
}
]
}'How to use Simon-Liu/gemma-4-e4b-github-mcp-sft-it with Docker Model Runner:
docker model run hf.co/Simon-Liu/gemma-4-e4b-github-mcp-sft-it
以 google/gemma-4-E4B-it 為基底,在 Simon-Liu/github-mcp-call-reasoning-1k 上做全參數監督式微調(full-parameter SFT),訓練成會先推理、再正確呼叫 GitHub MCP 工具(function calling)的繁體中文模型。
未微調基底 vs 微調後,在相同 held-out 樣本上的工具呼叫正確率:
| 指標 | FT 前 | FT 後 | 提升 |
|---|---|---|---|
| 有工具呼叫率 | 86.0% | 100.0% | +14.0% |
| 函式名正確率 | 86.0% | 100.0% | +14.0% |
| 名稱+參數全對率 | 74.0% | 98.0% | +24.0% |
有工具呼叫率=是否吐出
<tool_call>;函式名正確率=function name 正確;名稱+參數全對率=name 與 arguments 皆正確(exact match,最嚴格)。
google/gemma-4-E4B-itSimon-Liu/github-mcp-call-reasoning-1k(~1k 筆 GitHub MCP function-calling,繁中推理)from transformers import AutoModelForCausalLM, AutoTokenizer
import torch
model = AutoModelForCausalLM.from_pretrained("Simon-Liu/gemma-4-e4b-github-mcp-sft-it", torch_dtype=torch.bfloat16, device_map="auto")
tokenizer = AutoTokenizer.from_pretrained("Simon-Liu/gemma-4-e4b-github-mcp-sft-it")
messages = [
{"role": "system", "content": "<工具定義,Hermes 風格 JSON>"},
{"role": "user", "content": "我現在 GitHub 登入的帳號是誰?"},
]
prompt = tokenizer.apply_chat_template(messages, tokenize=False, add_generation_prompt=True)
inputs = tokenizer(prompt, return_tensors="pt").to(model.device)
out = model.generate(**inputs, max_new_tokens=512, do_sample=False)
print(tokenizer.decode(out[0][inputs["input_ids"].shape[1]:], skip_special_tokens=True))
衍生自 Google Gemma,使用前請遵守 Gemma 使用條款。