Instructions to use zhubolin/olmo2-0.837b-zh-en-sft with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use zhubolin/olmo2-0.837b-zh-en-sft with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("text-generation", model="zhubolin/olmo2-0.837b-zh-en-sft") messages = [ {"role": "user", "content": "Who are you?"}, ] pipe(messages)# Load model directly from transformers import AutoTokenizer, AutoModelForCausalLM tokenizer = AutoTokenizer.from_pretrained("zhubolin/olmo2-0.837b-zh-en-sft") model = AutoModelForCausalLM.from_pretrained("zhubolin/olmo2-0.837b-zh-en-sft", device_map="auto") 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
- llama.cpp
How to use zhubolin/olmo2-0.837b-zh-en-sft with llama.cpp:
Install (macOS, Linux)
curl -LsSf https://llama.app/install.sh | sh # Start a local OpenAI-compatible server with a web UI: llama serve -hf zhubolin/olmo2-0.837b-zh-en-sft:F16 # Run inference directly in the terminal: llama cli -hf zhubolin/olmo2-0.837b-zh-en-sft:F16
Install from WinGet (Windows)
winget install llama.cpp # Start a local OpenAI-compatible server with a web UI: llama serve -hf zhubolin/olmo2-0.837b-zh-en-sft:F16 # Run inference directly in the terminal: llama cli -hf zhubolin/olmo2-0.837b-zh-en-sft:F16
Use pre-built binary
# Download pre-built binary from: # https://github.com/ggerganov/llama.cpp/releases # Start a local OpenAI-compatible server with a web UI: ./llama-server -hf zhubolin/olmo2-0.837b-zh-en-sft:F16 # Run inference directly in the terminal: ./llama-cli -hf zhubolin/olmo2-0.837b-zh-en-sft:F16
Build from source code
git clone https://github.com/ggerganov/llama.cpp.git cd llama.cpp cmake -B build cmake --build build -j --target llama-server llama-cli # Start a local OpenAI-compatible server with a web UI: ./build/bin/llama-server -hf zhubolin/olmo2-0.837b-zh-en-sft:F16 # Run inference directly in the terminal: ./build/bin/llama-cli -hf zhubolin/olmo2-0.837b-zh-en-sft:F16
Use Docker
docker model run hf.co/zhubolin/olmo2-0.837b-zh-en-sft:F16
- LM Studio
- Jan
- vLLM
How to use zhubolin/olmo2-0.837b-zh-en-sft with vLLM:
Install from pip and serve model
# Install vLLM from pip: pip install vllm # Start the vLLM server: vllm serve "zhubolin/olmo2-0.837b-zh-en-sft" # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:8000/v1/chat/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "zhubolin/olmo2-0.837b-zh-en-sft", "messages": [ { "role": "user", "content": "What is the capital of France?" } ] }'Use Docker
docker model run hf.co/zhubolin/olmo2-0.837b-zh-en-sft:F16
- SGLang
How to use zhubolin/olmo2-0.837b-zh-en-sft 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 "zhubolin/olmo2-0.837b-zh-en-sft" \ --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": "zhubolin/olmo2-0.837b-zh-en-sft", "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 "zhubolin/olmo2-0.837b-zh-en-sft" \ --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": "zhubolin/olmo2-0.837b-zh-en-sft", "messages": [ { "role": "user", "content": "What is the capital of France?" } ] }' - Ollama
How to use zhubolin/olmo2-0.837b-zh-en-sft with Ollama:
ollama run hf.co/zhubolin/olmo2-0.837b-zh-en-sft:F16
- Unsloth Desktop
- Pi
How to use zhubolin/olmo2-0.837b-zh-en-sft with Pi:
Start the llama.cpp server
# Install llama.cpp: brew install llama.cpp # Start a local OpenAI-compatible server: llama serve -hf zhubolin/olmo2-0.837b-zh-en-sft:F16
Configure the model in Pi
# Install Pi: npm install -g @earendil-works/pi-coding-agent # Add to ~/.pi/agent/models.json: { "providers": { "llama-cpp": { "baseUrl": "http://localhost:8080/v1", "api": "openai-completions", "apiKey": "none", "models": [ { "id": "zhubolin/olmo2-0.837b-zh-en-sft:F16" } ] } } }Run Pi
# Start Pi in your project directory: pi
- Docker Model Runner
How to use zhubolin/olmo2-0.837b-zh-en-sft with Docker Model Runner:
docker model run hf.co/zhubolin/olmo2-0.837b-zh-en-sft:F16
- Lemonade
How to use zhubolin/olmo2-0.837b-zh-en-sft with Lemonade:
Pull the model
# Download Lemonade from https://lemonade-server.ai/ lemonade pull zhubolin/olmo2-0.837b-zh-en-sft:F16
Run and chat with the model
lemonade run user.olmo2-0.837b-zh-en-sft-F16
List all available models
lemonade list
- Hermes Agent
How to use zhubolin/olmo2-0.837b-zh-en-sft with Hermes Agent:
Start the llama.cpp server
# Install llama.cpp: brew install llama.cpp # Start a local OpenAI-compatible server: llama serve -hf zhubolin/olmo2-0.837b-zh-en-sft:F16
Configure Hermes
# Install Hermes: curl -fsSL https://hermes-agent.nousresearch.com/install.sh | bash hermes setup # Point Hermes at the local server: hermes config set model.provider custom hermes config set model.base_url http://127.0.0.1:8080/v1 hermes config set model.default zhubolin/olmo2-0.837b-zh-en-sft:F16
Run Hermes
hermes
- Atomic Chat
- OpenClaw
How to use zhubolin/olmo2-0.837b-zh-en-sft with OpenClaw:
Start the llama.cpp server
# Install llama.cpp: brew install llama.cpp # Start a local OpenAI-compatible server: llama serve -hf zhubolin/olmo2-0.837b-zh-en-sft:F16
Configure OpenClaw
# Install OpenClaw: npm install -g openclaw@latest # Register the local server and set it as the default model: openclaw onboard --non-interactive --mode local \ --auth-choice custom-api-key \ --custom-base-url http://127.0.0.1:8080/v1 \ --custom-model-id "zhubolin/olmo2-0.837b-zh-en-sft:F16" \ --custom-provider-id llama-cpp \ --custom-compatibility openai \ --custom-text-input \ --accept-risk \ --skip-health
Run OpenClaw
openclaw agent --local --agent main --message "Hello from Hugging Face"
olmo2-0.837b-zh-en-sft
从零训练的 0.837B 中英双语对话模型(SFT 版,未做偏好对齐)。纯个人学习项目。
模型信息
| 项目 | 值 |
|---|---|
| 架构 | OLMo2(reordered norm、QK-norm、tied embedding),16 层 / d_model 1536 / 16 头 / SwiGLU 6144 |
| 参数量 | 837,037,056(0.837B) |
| 上下文 | 4096 |
| Tokenizer | Qwen2.5(词表 151665),chat 模板为 Qwen 风格 |
| 预训练 | 132.1B tokens 单遍(中英约 3:7,悟道 2.0 + Dolma 1.7 子集),62,999 步,8×A100 |
| SFT | 509,515 条 / 129.7M tokens(中文 30.0%),3 epoch,lr 2e-5,4×A100 |
本仓库为 SFT 产物;DPO 对齐版见 zhubolin/olmo2-0.837b-zh-en-dpo(推荐使用)。
Chat 模板使用
<|im_start|>/<|im_end|>特殊标记包裹消息。
使用方法
from transformers import AutoModelForCausalLM, AutoTokenizer
path = "zhubolin/olmo2-0.837b-zh-en-sft"
tok = AutoTokenizer.from_pretrained(path)
model = AutoModelForCausalLM.from_pretrained(path, dtype="bfloat16", device_map="auto")
messages = [{"role": "user", "content": "保持健康的三个提示。"}]
text = tok.apply_chat_template(messages, tokenize=False, add_generation_prompt=True)
inputs = tok(text, return_tensors="pt").to(model.device)
out = model.generate(**inputs, max_new_tokens=512, do_sample=False)
print(tok.decode(out[0][inputs["input_ids"].shape[1]:], skip_special_tokens=True))
生成停止符:eos_token_id = [151645, 151643](<|im_end|> / <|endoftext|>)。
评测
OpenCompass(core 数据包子集,约 9.5K 题,SFT/DPO 同题对比):
局限性
- 0.8B 小模型:事实性知识薄弱,会产生幻觉,不要用于事实查询。
- 未做偏好/安全对齐,回答风格与拒答行为未调优。
- 纯语言数据训练,无代码、数学专项能力。
- 仅供学习交流。
- Downloads last month
- 907