
Anima
33B Chinese LLM, DPO QLORA, 100K context, AirLLM 70B inference with single 4GB GPU
Stars: 3380

Anima is the first open-source 33B Chinese large language model based on QLoRA, supporting DPO alignment training and open-sourcing a 100k context window model. The latest update includes AirLLM, a library that enables inference of 70B LLM from a single GPU with just 4GB memory. The tool optimizes memory usage for inference, allowing large language models to run on a single 4GB GPU without the need for quantization or other compression techniques. Anima aims to democratize AI by making advanced models accessible to everyone and contributing to the historical process of AI democratization.
README:
This is the first open source 33B Chinese LLM, we also support DPO alignment training and we have open source 100k context window. The latest update is AirLLM, a library helps you to infer 70B LLM from just single GPU with just 4GB memory.
第一个开源的基于QLoRA的33B中文大语言模型,支持了基于DPO的对齐训练。
我们也开源了100K输入窗口的开源模型Anima100K,基于Llama2,可商用。
最新开源了单卡跑70B模型的AirLLM。
Read this in English.
[2024/04/20] AirLLM supports Llama3 natively already. Run Llama3 70B on 4GB single GPU.
AirLLM天然支持Llama3 70B。4GB显存运行Llama3 70B大模型。
[2024/03/07] Open source: Latte text2video Training - Train your own SORA!
最接近SORA的开源模型来了!训练你自己的SORA
[2023/11/17] Open source: AirLLM, inference 70B LLM with 4GB single GPU.
开源AirLLM,单卡4GB显存跑70B大模型,无需量化,无需模型压缩
[2023/09/06] open source 100K context window Llama2 based LLM
更新支持100k 上下文的基于Llama2的可商用大模型
[2023/06/29] Open source alignment training based on DPO+QLORA
更新基于DPO+QLoRA的Human Feedback训练
[2023/06/12] Open source the first 33B Chinese Large language model
开源了第一个基于QLoRA的中文33B大语言模型
AirLLM optimizes inference memory usage, allowing 70B large language models to run inference on a single 4GB GPU card. No quantization, distillation, pruning or other model compression techniques that would result in degraded model performance are needed.
AirLLM优化inference内存,4GB单卡GPU可以运行70B大语言模型推理。不需要任何损失模型性能的量化和蒸馏,剪枝等模型压缩。
Find out more Here。
Train your own SORA:
Check out here: https://github.com/lyogavin/train_your_own_sora
We released the new Anima open source 7B model, supporting an input window length of 100K! It’s based on LLama2, so available for commercial use!
With specifically curated long text question answering training data for the 100K input length, and a lot of memory optimizations, we enabled the LLama2 model to scale to 100K input length.
当输入长度支持100k,你甚至可以把整个知识库都放入Prompt交给模型。或者可以把一本书直接放到Prompt里边。再也不用各种费劲的向量化,文本分割。。。。
我们堆了各种最新的猛料:XEntropy,Paged 8bit Adamw, LORA, Flashattention2,并且专门针对长输入对于training和Inference代码都做了修改定制,使得单卡100G就可以训练100k窗口。单卡40G就可以进行推理。
训练数据上,从几十种公开数据集中精选了专门针对长输入的30k~100k长度的长文本训练数据,专门针对100K输入对模型进行了训练。
Find out more Here。
We believe the future of AI will be fully open and democratized. AI should be a tool that’s accessible to everyone, instead of only the big monopolies(some of them have the term “open” in their names 😆 .). QLoRA might be an important step towards that future. We want to make some small contribution to the historical process of democratization of AI, we are open sourcing the 33B QLoRA model we trained: all the model parameters, code, datasets and evaluations are opened! 🤗
因此我们认为QLoRA 的工作很重要,重要到可能是个Game Changer。通过QLoRA的优化方法,第一次让33B规模的模型可以比较民主化的,比较低成本的finetune训练,并且普及使用。我们认为33B模型既可以发挥大规模模型的比较强的reasoning能力,又可以针对私有业务领域数据进行灵活的finetune训练提升对于LLM的控制力。
Find out more Here。
We open sourced latest alignment techinque - DPO.
Anima模型又开源了基于QLoRA的最新的DPO技术。
DPO是最新的最高效的RLHF训练方法。RLHF一直是生成式AI训练的老大难问题,也被认为是OpenAI的压箱底独家秘笈。DPO技术改变了这一切,让RLHF彻底傻瓜化!
我们开源了RLHF的低成本QLoRA的实现,一台GPU机器就可以训练33B模型的DPO!
Find out more here。
扫码:
扫码进群:
Buy me a coffee please! 欢迎大家参与贡献本项目 🙏
如果你喜欢我们的项目,请帮忙点个⭐吧!
This work is from Anima AI LLC and aiwrite.ai.
For Tasks:
Click tags to check more tools for each tasksFor Jobs:
Alternative AI tools for Anima
Similar Open Source Tools

Anima
Anima is the first open-source 33B Chinese large language model based on QLoRA, supporting DPO alignment training and open-sourcing a 100k context window model. The latest update includes AirLLM, a library that enables inference of 70B LLM from a single GPU with just 4GB memory. The tool optimizes memory usage for inference, allowing large language models to run on a single 4GB GPU without the need for quantization or other compression techniques. Anima aims to democratize AI by making advanced models accessible to everyone and contributing to the historical process of AI democratization.

Chat2DB
Chat2DB is an AI-driven data development and analysis platform that enables users to communicate with databases using natural language. It supports a wide range of databases, including MySQL, PostgreSQL, Oracle, SQLServer, SQLite, MariaDB, ClickHouse, DM, Presto, DB2, OceanBase, Hive, KingBase, MongoDB, Redis, and Snowflake. Chat2DB provides a user-friendly interface that allows users to query databases, generate reports, and explore data using natural language commands. It also offers a variety of features to help users improve their productivity, such as auto-completion, syntax highlighting, and error checking.

vertex-ai-mlops
Vertex AI is a platform for end-to-end model development. It consist of core components that make the processes of MLOps possible for design patterns of all types.

enchanted
Enchanted is an open-source, Ollama-compatible app for macOS and iOS that allows users to work with privately hosted models such as Llama 2, Mistral, Vicuna, Starling, and more. It provides a user-friendly interface for interacting with these models, making it easy to generate text, translate languages, write different kinds of creative content, and more. The app is designed to be secure and private, ensuring that user data is protected. It also offers a range of features such as dark/light mode, conversation history, markdown support, voice prompts, and image attachments.

sealos
Sealos is a cloud operating system distribution based on the Kubernetes kernel, designed for a seamless development lifecycle. It allows users to spin up full-stack environments in seconds, effortlessly push releases, and scale production seamlessly. With core features like easy application management, quick database creation, and cloud universality, Sealos offers efficient and economical cloud management with high universality and ease of use. The platform also emphasizes agility and security through its multi-tenancy sharing model. Sealos is supported by a community offering full documentation, Discord support, and active development roadmap.

Alpaca
Alpaca is an Ollama client for managing and chatting with multiple AI models. It offers a user-friendly way to interact with local AI and third-party providers like Gemini and ChatGPT. The open-source tool supports features such as multiple model conversations, image and document recognition, code highlighting, notifications, import/export chats, and more.

esp-ai
ESP-AI provides a complete AI conversation solution for your development board, including IAT+LLM+TTS integration solutions for ESP32 series development boards. It can be injected into projects without affecting existing ones. By providing keys from platforms like iFlytek, Jiling, and local services, you can run the services without worrying about interactions between services or between development boards and services. The project's server-side code is based on Node.js, and the hardware code is based on Arduino IDE.

semantic-router
The Semantic Router is an intelligent routing tool that utilizes a Mixture-of-Models (MoM) approach to direct OpenAI API requests to the most suitable models based on semantic understanding. It enhances inference accuracy by selecting models tailored to different types of tasks. The tool also automatically selects relevant tools based on the prompt to improve tool selection accuracy. Additionally, it includes features for enterprise security such as PII detection and prompt guard to protect user privacy and prevent misbehavior. The tool implements similarity caching to reduce latency. The comprehensive documentation covers setup instructions, architecture guides, and API references.

amazon-bedrock-client-for-mac
A sleek and powerful macOS client for Amazon Bedrock, bringing AI models to your desktop. It provides seamless interaction with multiple Amazon Bedrock models, real-time chat interface, easy model switching, support for various AI tasks, and native Dark Mode support. Built with SwiftUI for optimal performance and modern UI.

docling
Docling simplifies document processing, parsing diverse formats including advanced PDF understanding, and providing seamless integrations with the general AI ecosystem. It offers features such as parsing multiple document formats, advanced PDF understanding, unified DoclingDocument representation format, various export formats, local execution capabilities, plug-and-play integrations with agentic AI tools, extensive OCR support, and a simple CLI. Coming soon features include metadata extraction, visual language models, chart understanding, and complex chemistry understanding. Docling is installed via pip and works on macOS, Linux, and Windows environments. It provides detailed documentation, examples, integrations with popular frameworks, and support through the discussion section. The codebase is under the MIT license and has been developed by IBM.

FAV0
FAV0 Weekly is a repository that records weekly updates on front-end, AI, and computer-related content. It provides light and dark mode switching, bilingual interface, RSS subscription function, Giscus comment system, high-definition image preview, font settings customization, and SEO optimization. Users can stay updated with the latest weekly releases by starring/watching the repository. The repository is dual-licensed under the MIT License and CC-BY-4.0 License.

webots
Webots is an open-source robot simulator that provides a complete development environment to model, program, and simulate robots, vehicles, and mechanical systems. It was originally designed at EPFL in 1996 and further developed and commercialized by Cyberbotics since 1998. Webots was open-sourced in December 2018 and continues to be developed by Cyberbotics with paid customer support, training, and consulting services for industry and academic research projects.

tidb.ai
TiDB.AI is a conversational search RAG (Retrieval-Augmented Generation) app based on TiDB Serverless Vector Storage. It provides an out-of-the-box and embeddable QA robot experience based on knowledge from official and documentation sites. The platform features a Perplexity-style Conversational Search page with an advanced built-in website crawler for comprehensive coverage. Users can integrate an embeddable JavaScript snippet into their website for instant responses to product-related queries. The tech stack includes Next.js, TypeScript, Tailwind CSS, shadcn/ui for design, TiDB for database storage, Kysely for SQL query building, NextAuth.js for authentication, Vercel for deployments, and LlamaIndex for the RAG framework. TiDB.AI is open-source under the Apache License, Version 2.0.

EmbodiedScan
EmbodiedScan is a holistic multi-modal 3D perception suite designed for embodied AI. It introduces a multi-modal, ego-centric 3D perception dataset and benchmark for holistic 3D scene understanding. The dataset includes over 5k scans with 1M ego-centric RGB-D views, 1M language prompts, 160k 3D-oriented boxes spanning 760 categories, and dense semantic occupancy with 80 common categories. The suite includes a baseline framework named Embodied Perceptron, capable of processing multi-modal inputs for 3D perception tasks and language-grounded tasks.

ragstack-ai
RAGStack is an out-of-the-box solution simplifying Retrieval Augmented Generation (RAG) in GenAI apps. RAGStack includes the best open-source for implementing RAG, giving developers a comprehensive Gen AI Stack leveraging LangChain, CassIO, and more. RAGStack leverages the LangChain ecosystem and is fully compatible with LangSmith for monitoring your AI deployments.
For similar tasks

Anima
Anima is the first open-source 33B Chinese large language model based on QLoRA, supporting DPO alignment training and open-sourcing a 100k context window model. The latest update includes AirLLM, a library that enables inference of 70B LLM from a single GPU with just 4GB memory. The tool optimizes memory usage for inference, allowing large language models to run on a single 4GB GPU without the need for quantization or other compression techniques. Anima aims to democratize AI by making advanced models accessible to everyone and contributing to the historical process of AI democratization.

fsdp_qlora
The fsdp_qlora repository provides a script for training Large Language Models (LLMs) with Quantized LoRA and Fully Sharded Data Parallelism (FSDP). It integrates FSDP+QLoRA into the Axolotl platform and offers installation instructions for dependencies like llama-recipes, fastcore, and PyTorch. Users can finetune Llama-2 70B on Dual 24GB GPUs using the provided command. The script supports various training options including full params fine-tuning, LoRA fine-tuning, custom LoRA fine-tuning, quantized LoRA fine-tuning, and more. It also discusses low memory loading, mixed precision training, and comparisons to existing trainers. The repository addresses limitations and provides examples for training with different configurations, including BnB QLoRA and HQQ QLoRA. Additionally, it offers SLURM training support and instructions for adding support for a new model.

pgvecto.rs
pgvecto.rs is a Postgres extension written in Rust that provides vector similarity search functions. It offers ultra-low-latency, high-precision vector search capabilities, including sparse vector search and full-text search. With complete SQL support, async indexing, and easy data management, it simplifies data handling. The extension supports various data types like FP16/INT8, binary vectors, and Matryoshka embeddings. It ensures system performance with production-ready features, high availability, and resource efficiency. Security and permissions are managed through easy access control. The tool allows users to create tables with vector columns, insert vector data, and calculate distances between vectors using different operators. It also supports half-precision floating-point numbers for better performance and memory usage optimization.

TPI-LLM
TPI-LLM (Tensor Parallelism Inference for Large Language Models) is a system designed to bring LLM functions to low-resource edge devices, addressing privacy concerns by enabling LLM inference on edge devices with limited resources. It leverages multiple edge devices for inference through tensor parallelism and a sliding window memory scheduler to minimize memory usage. TPI-LLM demonstrates significant improvements in TTFT and token latency compared to other models, and plans to support infinitely large models with low token latency in the future.

KIVI
KIVI is a plug-and-play 2bit KV cache quantization algorithm optimizing memory usage by quantizing key cache per-channel and value cache per-token to 2bit. It enables LLMs to maintain quality while reducing memory usage, allowing larger batch sizes and increasing throughput in real LLM inference workloads.

bitsandbytes
bitsandbytes enables accessible large language models via k-bit quantization for PyTorch. It provides features for reducing memory consumption for inference and training by using 8-bit optimizers, LLM.int8() for large language model inference, and QLoRA for large language model training. The library includes quantization primitives for 8-bit & 4-bit operations and 8-bit optimizers.

pointer
Pointer is a lightweight and efficient tool for analyzing and visualizing data structures in C and C++ programs. It provides a user-friendly interface to track memory allocations, pointer references, and data structures, helping developers to identify memory leaks, pointer errors, and optimize memory usage. With Pointer, users can easily navigate through complex data structures, visualize memory layouts, and debug pointer-related issues in their codebase. The tool offers interactive features such as memory snapshots, pointer tracking, and memory visualization, making it a valuable asset for C and C++ developers working on memory-intensive applications.

qlora-pipe
qlora-pipe is a pipeline parallel training script designed for efficiently training large language models that cannot fit on one GPU. It supports QLoRA, LoRA, and full fine-tuning, with efficient model loading and the ability to load any dataset that Axolotl can handle. The script allows for raw text training, resuming training from a checkpoint, logging metrics to Tensorboard, specifying a separate evaluation dataset, training on multiple datasets simultaneously, and supports various models like Llama, Mistral, Mixtral, Qwen-1.5, and Cohere (Command R). It handles pipeline- and data-parallelism using Deepspeed, enabling users to set the number of GPUs, pipeline stages, and gradient accumulation steps for optimal utilization.
For similar jobs

weave
Weave is a toolkit for developing Generative AI applications, built by Weights & Biases. With Weave, you can log and debug language model inputs, outputs, and traces; build rigorous, apples-to-apples evaluations for language model use cases; and organize all the information generated across the LLM workflow, from experimentation to evaluations to production. Weave aims to bring rigor, best-practices, and composability to the inherently experimental process of developing Generative AI software, without introducing cognitive overhead.

LLMStack
LLMStack is a no-code platform for building generative AI agents, workflows, and chatbots. It allows users to connect their own data, internal tools, and GPT-powered models without any coding experience. LLMStack can be deployed to the cloud or on-premise and can be accessed via HTTP API or triggered from Slack or Discord.

VisionCraft
The VisionCraft API is a free API for using over 100 different AI models. From images to sound.

kaito
Kaito is an operator that automates the AI/ML inference model deployment in a Kubernetes cluster. It manages large model files using container images, avoids tuning deployment parameters to fit GPU hardware by providing preset configurations, auto-provisions GPU nodes based on model requirements, and hosts large model images in the public Microsoft Container Registry (MCR) if the license allows. Using Kaito, the workflow of onboarding large AI inference models in Kubernetes is largely simplified.

PyRIT
PyRIT is an open access automation framework designed to empower security professionals and ML engineers to red team foundation models and their applications. It automates AI Red Teaming tasks to allow operators to focus on more complicated and time-consuming tasks and can also identify security harms such as misuse (e.g., malware generation, jailbreaking), and privacy harms (e.g., identity theft). The goal is to allow researchers to have a baseline of how well their model and entire inference pipeline is doing against different harm categories and to be able to compare that baseline to future iterations of their model. This allows them to have empirical data on how well their model is doing today, and detect any degradation of performance based on future improvements.

tabby
Tabby is a self-hosted AI coding assistant, offering an open-source and on-premises alternative to GitHub Copilot. It boasts several key features: * Self-contained, with no need for a DBMS or cloud service. * OpenAPI interface, easy to integrate with existing infrastructure (e.g Cloud IDE). * Supports consumer-grade GPUs.

spear
SPEAR (Simulator for Photorealistic Embodied AI Research) is a powerful tool for training embodied agents. It features 300 unique virtual indoor environments with 2,566 unique rooms and 17,234 unique objects that can be manipulated individually. Each environment is designed by a professional artist and features detailed geometry, photorealistic materials, and a unique floor plan and object layout. SPEAR is implemented as Unreal Engine assets and provides an OpenAI Gym interface for interacting with the environments via Python.

Magick
Magick is a groundbreaking visual AIDE (Artificial Intelligence Development Environment) for no-code data pipelines and multimodal agents. Magick can connect to other services and comes with nodes and templates well-suited for intelligent agents, chatbots, complex reasoning systems and realistic characters.