ipex-llm
Large Language Model Develop Toolkit
Decision gist · record as of 2026-08-14
No. The project is archived and abandoned; Intel provides no maintenance, bug fixes, or support. Known security issues are documented and will not be patched. While the library may still function, adopting an unmaintained LLM acceleration tool introduces technical debt and security risk. Consider maintained alternatives or direct integration with actively supported frameworks.AI-flagged interpretation of the facts on this page — verify before relying
Before you install
- Requires Intel GPU (Arc, Flex, Max), NPU (Core Ultra), or compatible CPU; no runtime dependencies listed but underlying PyTorch and model libraries are implicit.
- Medium install friction with platform-specific wheels (manylinux2010 x86_64, Windows).
- Project is archived and abandoned as of the latest release; Intel will not provide maintenance, bug fixes, or support going forward.
License · maintenance · safety
Apache License, Version 2.0 (permissive) — Apache License 2.0 (permissive) allows commercial and private use with attribution; no restrictive copyleft obligations.
last release 2025-04-07 (494 days) · last repo commit 2026-01-28 · 8,862 stars · archived
0 known vulnerabilities (OSV.dev, 2026-08-14) · 322,364 downloads/mo, #7,612 on PyPI
Alternatives
Verify before relying
pip install ipex-llm
from ipex_llm.transformers import AutoModel
model = AutoModel.from_pretrained('model-name', load_in_4bit=True)- Exact scope of 70+ verified models and which specific versions are tested
- Whether archived status means security patches will be backported or if vulnerabilities remain unfixed
- Performance benchmarks for quantization modes (FP8/FP6/FP4/INT4) relative to unquantized baseline
- Compatibility matrix for specific Intel GPU models and driver versions required
What it is and what it does
IPEX-LLM is an acceleration library for running large language models on Intel hardware—GPUs (Arc, Flex, Max), NPUs (Core Ultra), and CPUs. It provides low-bit quantization (FP8, FP6, FP4, INT4) and integrates with popular frameworks like PyTorch, llama.cpp, Ollama, vLLM, LangChain, and HuggingFace transformers to reduce memory footprint and improve inference speed on Intel-based systems.
The library supports over 70 models including Llama, Phi, Mistral, Qwen, and DeepSeek variants. However, the project is now archived and abandoned—Intel will not provide maintenance, bug fixes, new releases, or support. The description notes known security issues exist. Users should treat this as a snapshot tool rather than an actively maintained dependency.
Use it for
- Run quantized LLMs locally on Intel Arc GPUs without external cloud services
- Integrate low-bit quantized models into Ollama or llama.cpp workflows on Intel hardware
- Serve LLMs via vLLM with Intel GPU acceleration for batch inference
- Fine-tune models using Axolotl with Intel GPU backend
- Deploy multimodal models (Qwen-VL, Phi-3-Vision) on Intel NPU or GPU
Worth the install?
AI-flagged interpretation of the facts on this page. Verify before relying on it.
No.
The project is archived and abandoned; Intel provides no maintenance, bug fixes, or support. Known security issues are documented and will not be patched. While the library may still function, adopting an unmaintained LLM acceleration tool introduces technical debt and security risk. Consider maintained alternatives or direct integration with actively supported frameworks.
Install
ipex-llm on PyPI
Before you install
Medium install friction with platform-specific wheels (manylinux2010 x86_64, Windows). Project is archived and abandoned as of the latest release; Intel will not provide maintenance, bug fixes, or support going forward.
Requires Intel GPU (Arc, Flex, Max), NPU (Core Ultra), or compatible CPU; no runtime dependencies listed but underlying PyTorch and model libraries are implicit.
License in practice
Apache License 2.0 (permissive) allows commercial and private use with attribution; no restrictive copyleft obligations.
Quickstart
pip install ipex-llm
from ipex_llm.transformers import AutoModel
model = AutoModel.from_pretrained('model-name', load_in_4bit=True)Verify before relying
- Exact scope of 70+ verified models and which specific versions are tested
- Whether archived status means security patches will be backported or if vulnerabilities remain unfixed
- Performance benchmarks for quantization modes (FP8/FP6/FP4/INT4) relative to unquantized baseline
- Compatibility matrix for specific Intel GPU models and driver versions required
Package facts
| License | Apache License, Version 2.0 permissive |
| Python support | Not specified |
| Install friction | Medium. Platform-specific wheel |
| Runtime dependencies | None |
| Maintenance | Abandoned 494 days since the last release |
| Last repo commit | repository archived |
| First released | |
| Downloads | 322,364 / month, #7,612 on PyPI 30-day window, as of 2026-08-14 |
| Known vulnerabilities | None known OSV.dev, checked 2026-08-14 |
| Classifiers | License :: OSI Approved :: Apache Software LicenseProgramming Language :: Python :: 3Programming Language :: Python :: 3.11Programming Language :: Python :: Implementation :: CPython |
Evidence: ipex_llm-2.2.0-py3-none-manylinux2010_x86_64.whl; ipex_llm-2.2.0-py3-none-win_amd64.whl
Tags
Let your AI agent find packages like this
Example. Real query, live index.
You found this page by searching. An agent finds it by wishing: SkillFed indexes 14,416 PyPI packages by what they can do, searchable in plain language.
wish › “llm acceleration intel gpu”
- ipex-llmAccelerates large language model inference on Intel hardware (GPU,…
- torch-directmlEnables PyTorch to use DirectML for GPU acceleration on DirectX…
- scikit-learn-intelexAccelerates scikit-learn algorithms on CPU and GPU by patching or…
Give your agent the search over MCP, or paste the wish link into any chat.
More Artificial Intelligence packages
LiteLLM provides a unified Python interface to call 100+ LLM providers (OpenAI, Anthropic, Gemini, Bedrock, Azure, and others) using OpenAI-compatible API format, available as both a Python SDK and a self-hosted AI Gateway proxy server.
Install it if you need to work with multiple LLM providers or want to centralize LLM routing in your organization.
Client library and CLI tool for downloading, uploading, and managing models, datasets, and repositories on the Hugging Face Hub platform.
Install it if you work with Hugging Face Hub models or datasets.
LangChain provides a framework for building agents and LLM-powered applications by composing language models, tools, and memory through a unified API that abstracts over multiple model providers.
hf-xet provides chunk-based deduplication and efficient file transfer for the Hugging Face Hub, enabling faster uploads and downloads of large files with local disk caching.
Tokenizers converts raw text into token sequences for NLP models, with support for training custom vocabularies and using pre-built tokenizers (BPE, WordPiece) optimized for speed via Rust.
Transformers provides a unified framework for loading, fine-tuning, and running state-of-the-art pretrained models across text, vision, audio, video, and multimodal tasks using PyTorch, JAX, or TensorFlow.
Install it if you need to run or train any transformer-based model for NLP, vision, audio, or multimodal tasks.
See also abstract-hugpy-dev · vllm · vllm-cpu · nvidia-modelopt · vllm-tpu · llm · llama-cpp-python · lm-eval · unsloth · auto-round