GLM-5.2 is live. #1 throughput on OpenRouter, pay-per-token on FriendliAI. Try it today ➜
coslinedev
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
coslinedev
Model Tree
Qwen/Qwen2.5-3B-Instruct
Input Modalities
Output Modalities
Supported Functionality
This repository contains the full merged model weights (~6.18 GB), ready for direct inference without requiring extra LoRA adapter layers:
model-00001-of-00002.safetensors & model-00002-of-00002.safetensors: Full model weight shards.model.safetensors.index.json: Index mapping for sharded weight loading.chat_template.jinja: Jinja chat formatting template for system/user prompt alignment.tokenizer.json & tokenizer_config.json: Tokenizer assets inherited from Qwen2.5.config.json & generation_config.json: Model architecture and generation configuration parameters.You can load and run this model locally or on Google Colab using Hugging Face transformers:
python
import torchfrom transformers import AutoModelForCausalLM, AutoTokenizerMODEL_ID = "coslinedev/Qwen2.5-3B-FinCode-Reasoning-Full"tokenizer = AutoTokenizer.from_pretrained(MODEL_ID, trust_remote_code=True)model = AutoModelForCausalLM.from_pretrained(MODEL_ID,torch_dtype=torch.float16,device_map="auto")prompt = "Write a Python function to calculate Double Declining Balance (DDB) depreciation with a salvage floor constraint."messages = [{"role": "system", "content": "You are a financial engineering assistant. Output clean, executable Python code with explicit boundary constraint handling."},{"role": "user", "content": prompt}]formatted_prompt = tokenizer.apply_chat_template(messages, tokenize=False, add_generation_prompt=True)inputs = tokenizer([formatted_prompt], return_tensors="pt").to(model.device)outputs = model.generate(**inputs, max_new_tokens=1024, temperature=0.2)print(tokenizer.decode(outputs[0][inputs.input_ids.shape[1]:], skip_special_tokens=True))⚙️ Key Design PrinciplesExecution Sandbox First: Direct numerical output is delegated to the Python interpreter executing the generated script, eliminating calculation hallucinations.Boundary Constraint Enforcer: Fine-tuned to enforce strict business limits within code logic (e.g., depreciation = min(book_value * ddb_rate, max(book_value - salvage, 0.0))).Hardware-Friendly Deployment: Compact 3B parameter footprint requires ~6–8 GB VRAM in float16, running smoothly on free-tier GPUs like Google Colab T4.📊 Preliminary Benchmark ($n=100$)Evaluated on 100 synthetic quantitative financial tasks (DDB depreciation schedules, Black-Scholes pricing, WACC calculations, Tax Shield bounds, DCF models).Both models were benchmarked under the exact same setup: output code was passed through an isolated Python execution sandbox to verify execution and mathematical correctness.ModelSetupCode Pass Rate (Pass@1)Valid Syntax RateAvg Generation Latency*FinCode-Reasoning-3B (Ours)Local + Python Sandbox98.0%99.5%0.85sQwen2.5-3B-Instruct (Base)Local + Python Sandbox82.0%85.0%0.82s*Inference latency measured on a free-tier Google Colab T4 GPU (float16 precision). Does not include sandbox execution overhead.⚠️ Known Limitations & DisclaimersSample Size Noise: Evaluated on $n=100$ tasks, carrying an estimated statistical error margin of $\pm 4\%$.Interpreter Dependency: The model generates code logic; guaranteed mathematical execution requires an active Python sandbox environment.Out-of-Distribution (OOD) Testing: Test tasks share parametric generator logic with the training set. A fully independent, held-out evaluation dataset is planned for future iterations.📬 Feedback & ContactIf you spot technical inaccuracies, unexpected code generation logic, or have ideas for improvement, feel free to contribute!Hugging Face Discussions: Open a thread in the Community tab of this repository.Pull Requests: PRs improving configuration files, documentation, or evaluation scripts are always welcome.📜 LicenseLicensed under the Apache 2.0 License, aligning with base Qwen2.5 model licensing.