Ovis2
Alibaba's open source multimodal large language model with powerful visual understanding, OCR, video processing and reasoning capabilities, supporting multiple scale versions.
SenseNovaLarge ModelIt is a comprehensive big modeling system launched by ShangTech.
Background & Release:
- Day by day new SenseNova is a big modeling system announced by ShangTech in April 2023 by Chairman and CEO Xu Li.
- The system was approved to go live in August 2023, marking its official availability to the public.
- On May 29th, 2024, ShangTech announced that its "DayDayNew" Big Model will soon undergo a major upgrade, and officially released the DayDayNew Big Model 5.0 Cantonese version to the public.
Main Functions and Features::
- natural language processing (NLP): Automatically transforms data into meaningful analytics and visualization results through a combination of code generation and automated execution through capabilities such as natural language generation, intent recognition, logic understanding and code interpreters.
- Vincentian graphic ability (geology):: Includesdigital personThe video generation platform "SenseAvatar" (SenseAvatar) and other features can provide users with rich visual content generation services.
- Model Development Functions: Support users to develop models according to their needs and provide customized AI solutions.
Technical characteristics::
- Hybrid Expert Architecture (MoE): Nisshin SenseNova 5.0 utilizes the MOE Hybrid Expert Architecture, an architecture that allows the model to complete inference with a small number of parameters activated, improving the model's processing efficiency and responsiveness.
- Volume of training data: Based on more than 10TB tokens of training data, it ensures that the model has a robust knowledge base and a wide range of applications.
- inference context window: Reaching around 200K allows the model to handle longer text sequences and more complex contextual relationships.
Performance benchmarking::
Rizhixin SenseNova 5.0 fully benchmarks the GPT-4 Turbo in terms of comprehensive performance and meets or exceeds the GPT-4 Turbo in mainstream objective reviews, especially in terms of natural language capability, text-to-graph capability, multimodal and data analysis capability.
application scenario::
Risen SenseNova big models have been widely used in many fields such as finance, healthcare, education, etc., such as intelligent customer service, intelligent marketing, investment research analysis, research report writing, medical and healthcare language big models, etc., which provide powerful AI support for the industry.
language version::
In addition to the standard version, ShangTech has also released the Day Day New Big Model 5.0 Cantonese version, which further extends the model's language support capabilities.
Rizhixin SenseNova Big Model is a comprehensive big model system launched by ShangTech, with powerful natural language processing, text-born graph capabilities and model development functions. Through advanced hybrid expert architecture and a large amount of training data, it realizes performance comparable to GPT-4 Turbo, and is widely used in a variety of fields, providing users with efficient and flexible AI services.
data statistics
Relevant Navigation
DeepSeek-V4
The new generation of domestic open-source flagship big model has become one of the strongest all-around AIs on the ground with millions of ultra-long contexts, performance comparable to the top international closed-source models, and extreme cost-effectiveness.
Qwen3-Next
Ali open source 80 billion parameters of the big model, 1:50 super sparse activation, millions of contexts, the cost down 90%, the performance is comparable to the hundreds of billions of models.
Qwen 3.6-Plus
Alibaba launched the strongest domestic programming model, with intelligent body programming, multimodal reasoning and millions of contexts as the core, support for the independent disassembly and execution of complex tasks, taking into account the high performance and low threshold, known as the developers and enterprises of the “all-round programming assistant”.
GWM-1
Runway's first universal world model simulates physical laws and dynamic environments through frame-by-frame pixel prediction technology. It supports robot training, digital human generation, and cross-domain simulation, redefining how AI understands and interacts with the world.
Zidong Taichu
The cross-modal general artificial intelligence platform developed by the Institute of Automation of the Chinese Academy of Sciences has the world's first graphic, text and audio three-modal pre-training model with cross-modal comprehension and generation capabilities, supporting full-scene AI applications, which is a major breakthrough towards general artificial intelligence.
QwQ-32B
Alibaba released a high-performance inference model with 32 billion parameters that excels in mathematics and programming for a wide range of application scenarios.
Xiaomi MiMo
Xiaomi's open-sourced 7 billion parameter inference macromodel, which outperforms models such as OpenAI o1-mini in mathematical reasoning and code competitions by a small margin.
No comments...
简体中文 
English