GLM-5.2 is live. #1 throughput on OpenRouter, pay-per-token on FriendliAI. Try it today ➜
MSGEncrypted
Available on FriendliAI
Run this model inference on single tenant GPU with unmatched speed and reliability at scale.
Run this model inference with full control and performance in your environment.
Model Details
Model Provider
MSGEncrypted
Model Tree
openbmb/MiniCPM5-1B
Input Modalities
Output Modalities
Supported Functionality
This repository contains a math-focused PEFT LoRA adapter based on [openbmb/MiniCPM5-1B](https://huggingface.co/ openbmb/MiniCPM5-1B).
openbmb/MiniCPM5-1Bloraadapter32645.0150This model is intended for research and experimentation with math reasoning fine-tuning. It is not intended for high-stakes educational, financial, legal, medical, or safety-critical decisions.
The adapter was trained on math reasoning data, including MetaMathQA and Hendrycks MATH-style algebra/geometry examples, depending on the notebook configuration used for this run.
The run used lora fine-tuning with PEFT/LoRA-style adaptation where applicable. Training configuration and adapter metadata are included in adapter_enhancement_metadata.json.
Before/after evaluation was run with the same base model and PEFT disable_adapter() for the baseline.
| Task | Role | N | Before | After | Delta |
|---|---|---|---|---|---|
| GSM8K | primary | 30 | 0.133 | 0.200 | +0.067 |
| PIQA | guard | 150 | 0.620 | 0.647 | +0.027 |
Gate status: PASS.
python
from transformers import AutoModelForCausalLM, AutoTokenizerfrom peft import PeftModelbase_model = "openbmb/MiniCPM5-1B"adapter_id = "MSGEncrypted/minicpm5-1b-math-lora"tokenizer = AutoTokenizer.from_pretrained(base_model, trust_remote_code=True)base = AutoModelForCausalLM.from_pretrained(base_model,torch_dtype="auto",device_map="auto",trust_remote_code=True,)model = PeftModel.from_pretrained(base, adapter_id)prompt = "Solve: If there are 12 apples and 5 are eaten, how many remain?"inputs = tokenizer(prompt, return_tensors="pt").to(model.device)outputs = model.generate(**inputs, max_new_tokens=128, do_sample=False)print(tokenizer.decode(outputs[0], skip_special_tokens=True))
The model may produce incorrect reasoning, hallucinated steps, or invalid arithmetic. Evaluation results from small notebook samples should be treated as smoke tests, not full benchmark claims.
eval_results.json: captured GSM8K/PIQA notebook resultsadapter_enhancement_metadata.json: training and publish metadata2.10.0+cu128