With only 2.7 billion parameters, Microsoft releases a new Phi-2 model! - Code World

With only 2.7 billion parameters, Microsoft releases a new Phi-2 model!

Enterprise 2023-12-17 17:38:52 views: null

NoSuchKey

Guess you like

Origin blog.csdn.net/dQCFKyQDXYm3F8rB0/article/details/135007434

With only 2.7 billion parameters, Microsoft releases a new Phi-2 model!

There are only 2.7 billion parameters, but the performance is 25 times higher! Microsoft releases Phi-2

There are only 2.7 billion parameters, but the performance is 25 times higher! Microsoft releases Phi-2

There are only 2.7 billion parameters, but the performance is 25 times higher! Microsoft releases Phi-2

There are only 2.7 billion parameters, but the performance is 25 times higher! Microsoft releases Phi-2

There are only 2.7 billion parameters, but the performance is 25 times higher! Microsoft releases Phi-2

There are only 2.7 billion parameters, but the performance is 25 times higher! Microsoft releases Phi-2

There are only 2.7 billion parameters, but the performance is 25 times higher! Microsoft releases Phi-2

Microsoft launches small model Phi-2 with better performance than Llama 2/Mistral 7B

Mistral AI releases Mistral 7B, a model with 7.3 billion parameters

Breaking language barriers: Google releases the M4 translation model with 50 billion training parameters and supports 103 languages

Mistral AI releases 7.3 billion parameter model, "crushing" Llama 2 13B

3.6 trillion tokens, 340 billion parameters, details of Google's large model PaLM 2 exposed

Benchmark Gen-2! Meta releases new model and enters Vincent video track

Fudan Qiu Xipeng's new work: single-machine fine-tuning of a large model with 65 billion parameters, industry insiders: it is of great significance to the popularization of large models...

Tencent Tang Daosheng: With over 100 billion parameters and over 2 trillion tokens, Tencent’s Hunyuan large model is fully open to the industry

Releases new model class package (for NuGet uploaded to the server)

Orca: Microsoft tests new AI model

Google LaMDA large language model releases a new application, crushing ChatGPT and causing a boom, breaking through 2 million installations within a week

PyTorch saves some model parameters and loads them in a new model

Microsoft's open source depth study optimized libraries DeepSpeed, trainable 100 billion parameter model

ZeRO & DeepSpeed: allows training model has more than 100 billion parameter optimization (Microsoft)

A large model with 7 billion parameters running on the iPhone, the latest achievement from Chen Tianqi's team

Wang Xiaochuan's big model debut! 7 billion parameters dominate the list, and Qingbei is the first to use it｜Exclusive interview

Introduction to LLaMA: An introduction to the official website of a large-scale language model with 65 billion parameters

Abandoning Softmax, the first large linear attention Transformer model: 175 billion parameters, better speed and accuracy

I installed a large model with 7 billion parameters on the iPhone, the latest achievement from Chen Tianqi's team

180 billion parameters, supports Chinese, 3.5 trillion training data! Open source ChatGPT-like model

Celebrate Windows 10 users to break 1 billion, Microsoft demonstrated the new UI design

Burst! Microsoft's new work LongNet: Extend Transformer to 1 billion Tokens

Recommended

Ranking

Followed Deng data structure of the 1-a (Introduction)

Navedi made a wonderful appearance at the 10th DCD Beijing Data Center Conference, boosting industry development

DataNode offline speed optimization

[In-depth understanding of JVM]: ClassLoader (ClassLoader) and parent delegation model

Flex learning summary

[Java] traverse Map <String, String>

WeChat red envelope algorithm

For the digital economy to take off in the park, it must first grow "network wings"

Analysis of Reactor Thread Model

LSTM model theoretical summary (generation, development and performance, etc.)

Daily

More

2025-03-03(0)

2025-03-02(0)

2025-03-01(0)

2025-02-28(0)

2025-02-27(0)

2025-02-26(0)

2025-02-25(0)

2025-02-24(0)

2025-02-23(0)

2025-02-22(0)