Downloads · 30 days
10
1% of all-time downloads
Irfanuruchi/qwen2.5-3b-buildeng-openvino-fp16
qwen2.5-3b-buildeng-openvino-fp16 is a text generation model from Irfanuruchi. Use it when you need the model to write or continue text. It is set up for openvino. The card lists the license as other.
BuildEng V8 3B OpenVINO FP16 is the higher-precision OpenVINO release of BuildEng V8 3B, a domain-specialized language model focused on civil engineering, structural reasoning, construction workflows, and conservative…
Downloads · 30 days
10
1% of all-time downloads
All-time downloads
669
Public
Repo size
6.2 GB
Likes
1
Public
Click a slice to open those files.
.bin6.2 GB · 100%
From the Hugging Face model README
BuildEng V8 3B OpenVINO FP16 is the higher-precision OpenVINO release of BuildEng V8 3B, a domain-specialized language model focused on civil engineering, structural reasoning, construction workflows, and conservative engineering-assistant behavior.
This release is converted from:
Irfanuruchi/qwen2.5-3b-buildengThe BuildEng V8 3B model is fine-tuned from:
Qwen/Qwen2.5-3B-InstructImproved using Qwen.
| Property | Value |
|---|---|
| Model | BuildEng V8 3B |
| Runtime | OpenVINO |
| Precision | FP16 |
| BuildEng source model | Irfanuruchi/qwen2.5-3b-buildeng |
| Upstream model | Qwen/Qwen2.5-3B-Instruct |
| Domain | Civil and structural engineering |
| Primary use | Intel-optimized local engineering-assistant inference |
This FP16 release is intended for users who prioritize model precision and output quality over reduced storage or memory requirements.
BuildEng is designed around engineering and construction workflows including:
The model is designed to function as an engineering assistant and to avoid presenting uncertain conclusions as established facts.
The OpenVINO FP16 release is intended to provide:
A 3B INT4 OpenVINO export was evaluated during development but was not published because its output was not sufficiently stable for release.
pip install "optimum-intel[openvino]" openvino transformers
from optimum.intel.openvino import OVModelForCausalLM
from transformers import AutoTokenizer
model_id = "Irfanuruchi/qwen2.5-3b-buildeng-openvino-fp16"
tokenizer = AutoTokenizer.from_pretrained(model_id)
model = OVModelForCausalLM.from_pretrained(
model_id,
device="CPU",
compile=True,
)
messages = [
{
"role": "system",
"content": (
"You are BuildEng. Give concise, inspection-first "
"building engineering guidance."
),
},
{
"role": "user",
"content": (
"A homeowner reports diagonal cracks near a window corner. "
"What should be inspected first?"
),
},
]
prompt = tokenizer.apply_chat_template(
messages,
tokenize=False,
add_generation_prompt=True,
)
inputs = tokenizer(prompt, return_tensors="pt")
output = model.generate(
**inputs,
max_new_tokens=160,
do_sample=False,
)
new_tokens = output[0][inputs["input_ids"].shape[-1]:]
print(tokenizer.decode(new_tokens, skip_special_tokens=True))
Depending on the OpenVINO installation, hardware, and platform support, this release may be used with:
Irfanuruchi/qwen2.5-3b-buildeng| Format | Repository |
|---|---|
| Hugging Face | Irfanuruchi/qwen2.5-3b-buildeng |
| GGUF Q4_K_M | Irfanuruchi/qwen2.5-3b-buildeng-GGUF-Q4_K_M |
| GGUF Q8_0 | Irfanuruchi/qwen2.5-3b-buildeng-GGUF-Q8_0 |
| GGUF F16 | Irfanuruchi/qwen2.5-3b-buildeng-GGUF-F16 |
| MLX 4-bit | Irfanuruchi/qwen2.5-3b-buildeng-mlx-4bit |
| MLX 8-bit | Irfanuruchi/qwen2.5-3b-buildeng-mlx-8bit |
| OpenVINO INT8 | Irfanuruchi/qwen2.5-3b-buildeng-openvino-int8 |
Irfanuruchi/buildeng-v8-3bBuildEng V8 3B is derived from Qwen/Qwen2.5-3B-Instruct. The upstream Qwen2.5-3B-Instruct model is distributed under the Qwen Research License Agreement, and the applicable upstream licensing terms continue to apply to this OpenVINO derivative release.
Under the upstream Qwen Research License Agreement, the default grant of rights to use, reproduce, distribute, modify, and create derivative works from the Qwen materials is limited to non-commercial research and evaluation purposes.
Commercial use of the upstream Qwen materials requires requesting the appropriate license from Alibaba Cloud.
Accordingly, Apache-2.0 licensing applicable to other BuildEng model families must not be interpreted as granting Apache-2.0 rights over the upstream Qwen2.5-3B materials contained in this release.
A copy of the upstream Qwen Research License Agreement is included in this repository as:
LICENSEThe required Qwen attribution and BuildEng modification information are included in:
NOTICEQwen is licensed under the Qwen RESEARCH LICENSE AGREEMENT, Copyright (c) Alibaba Cloud. All Rights Reserved.
Improved using Qwen.
For applications that require a BuildEng model based on an upstream Apache-2.0-licensed model, BuildEng variants based on Qwen2.5-1.5B and Qwen2.5-32B are available separately.
Users are responsible for reviewing and complying with all licenses applicable to their intended use and deployment.
BuildEng is an engineering-assistant model and is not a substitute for professional engineering judgment.
Model output should not be treated as:
Engineering decisions should be independently verified using appropriate calculations, standards, drawings, site information, project-specific constraints, and professional review.
This repository contains an OpenVINO FP16 conversion of the fine-tuned BuildEng V8 3B model.
Model lineage:
Qwen/Qwen2.5-3B-Instruct
→ Irfanuruchi/qwen2.5-3b-buildeng
→ Irfanuruchi/qwen2.5-3b-buildeng-openvino-fp16
BuildEng-specific model development, fine-tuning, engineering specialization, documentation, and related project contributions were produced by Irfan Uruçi.
The OpenVINO FP16 conversion represented by this repository is a modified/converted derivative of the BuildEng V8 3B release.
Upstream Qwen intellectual-property rights remain subject to the Qwen Research License Agreement.
Irfan Uruçi BuildEng Project