Langboat Cognitive Intelligence Platform
Langboat Cognitive Intelligence Platform is an industry knowledge service cloud platform with NLP as the core. It provides a series of services such as search, generation, translation, and dialogue.
Enterprise Applications
Langboat Smart Knowledge Base
An enterprise intelligent knowledge base platform in the era of large models.
Langboat Smart Agent Builder
Enterprise Intelligent Agent Application Development Tool Platform
Langboat Meeting Assistant Platform
Deep analysis of meeting audio and video, quickly capturing key information.
Langboat Intelligent Translation
Langboat's intelligent translation matrix, leading in effectiveness, with flexible and diverse usage.
Mengzi Models
A powerful self-developed generative controllable large language model.
Intelligent Enterprise Knowledge Base Solution
Intelligent Customer Service Solution
Chat BI SolutionMengzi GPT
The Mengzi GPT Model is a large-scale language model developed with Langboat self-developed technology. It can handle multilingual and multimodal data, while supporting multiple text comprehension and generation tasks, meeting the needs of different fields and application scenarios. Mengzi is a well-known domestic model brand that has achieved excellent results in benchmark evaluations of Chinese LLMs, such as C-EVAL and SUPERCLUE. The Mengzi model has been registered with the China Cyberspace Administration for Generative Artificial Intelligence by the end of 2023, and it has officially opened to the public for general service.
Product Overview
Knowledge Q&A
Helping users get the information they need directly through Q&A, eliminating the need for searching and filtering
General Writing
Ability to write articles on a variety of genres and topics according to user requirements and control the effectiveness of content creation through natural language dialogue
Multilingual Translation
Multilingual translation in dialogue, more fluent and natural than traditional translation
Financial Scenario Tasks
Optimizing models for special tasks in financial scenarios to help industry workers complete their work more efficiently
Document Q&A
Retrieve answers from given documents and respond to user queries
Code Generation
Automatically generate programming code snippets based on user requirements
Technical Solutions
Support Multiple Model Architectures
- Autoregressive models: such as GPT
- Self-encoding models: such as BERT
- Encoder-Decoder model: T5
Lightweight Model Performance Enhancement
- Fusion of multiple pre-training tasks
- SMART adverserial training
- Knowledge distillation
Knowledge Graph Based Enhancement
- Enhancements with entity extraction
- Knowledge graph enhancement (isa relationship)
- Knowledge graph to text conversion
Linguistic Knowledge Based Enhancement
- Mask mechanism enhanced by syntactic information
- Semantic role embedding enhancement
- Attention weight constrained pruning of dependencies
Few-Shot/Zero-Shot Learning
- Prompt template construction
- Multi-task learning technique
- Common information extraction scenarios, out of the box
Retrieval Based Enhancement
- Knowledge decoupling
- Strong interpretability
- External knowledge components are updated in real time
Technical Advantages
It has achieved better performance than conventional models in multiple tasks
It supports BERT, GPT, T5 and other architectures, with different scenarios covered
It supports image and text dual-mode input, which better handles image and text related tasks
It supports rapid optimization for vertical domains, and offers models scaling from 10M to 1B parameters
Capability and Qualification
C-Eval Leaderboard
*Ranking as of August, 2023
| # | 0 | 1 | 2 | 3 | 4 |
|---|---|---|---|---|---|
| Model | Mengzi | ChatGLM2 | InternLM-123B | GPT-4* | AiLMe-100B v2 |
| Creator | Langboat | Tsinghua & Zhipi.AI | Shanghai AI Lab & Sense Time | OpenAI | APUS |
| Submission Date | 2023/8/25 | 2023/6/25 | 2023/8/22 | 2023/5/15 | 2023/7/25 |
| Avg | 71.5 | 71.1 | 68.8 | 68.7 | 67.7 |
| Avg(Hard) | 48.8 | 50 | 50 | 54.9 | 55.3 |
| STEM | 62.3 | 64.4 | 63.5 | 67.1 | 65.4 |
| Social Science | 87.2 | 81.6 | 81.4 | 77.6 | 72.3 |
| Humanities | 76.8 | 73.7 | 72.7 | 64.5 | 71.2 |
| Others | 68.6 | 71.3 | 63 | 67.8 | 64 |
| # | Model | Creator | Submission Date | Avg | Avg(Hard) | STEM | Social Science | Humanities | Others |
|---|---|---|---|---|---|---|---|---|---|
| 0 | Mengzi | Langboat | 2023/8/25 | 71.5 | 48.8 | 62.3 | 87.2 | 76.8 | 68.6 |
| 1 | ChatGLM2 | Tsinghua & Zhipi.AI | 2023/6/25 | 71.1 | 50 | 64.4 | 81.6 | 73.7 | 71.3 |
| 2 | InternLM-123B | Shanghai AI Lab & Sense Time | 2023/8/22 | 68.8 | 50 | 63.5 | 81.4 | 72.7 | 63 |
| 3 | GPT-4* | OpenAI | 2023/5/15 | 68.7 | 54.9 | 67.1 | 77.6 | 64.5 | 67.8 |
| 4 | AiLMe-100B v2 | APUS | 2023/7/25 | 67.7 | 55.3 | 65.4 | 72.3 | 71.2 | 64 |
CLUE Leaderboards
*Ranking as of July 30, 2021
| Ranking | 1 | 2 | 3 | |
|---|---|---|---|---|
| Model | Mengzi | Motian | BETRTSG | Human Level |
| Scale | 1B | 1B | 10B | |
| Total Score | 82.90 | 82.15 | 81.80 | 86.68 |
| AFQMC | 79.82 | 78.30 | 79.85 | 81.00 |
| TNEWS | 64.68 | 57.42 | 57.42 | 71.00 |
| IFLYTEK | 65.08 | 65.46 | 64.54 | 80.30 |
| OCNLI | 81.87 | 84.97 | 85.93 | 90.30 |
| WSC2020 | 96.55 | 94.83 | 95.17 | 98.00 |
| CSL | 89.87 | 90.17 | 89.00 | 84.00 |
| CMRC2018 | 82.25 | 85.30 | 83.80 | 92.40 |
| CHID | 96.00 | 94.43 | 93.06 | 87.10 |
| C3 | 89.98 | 88.49 | 87.44 | 96.00 |
| Ranking | Model | Scale | Total Score | AFQMC | TNEWS | IFLYTEK | OCNLI | WSC2020 | CSL | CMRC2018 | CHID | C3 |
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 1 | Mengzi | 1B | 82.90 | 79.82 | 64.68 | 65.08 | 81.87 | 96.55 | 89.87 | 82.25 | 96.00 | 89.98 |
| 2 | Motian | 1B | 82.15 | 78.30 | 57.42 | 65.46 | 84.97 | 94.83 | 90.17 | 85.30 | 94.43 | 88.49 |
| 3 | BETRTSG | 10B | 81.80 | 79.85 | 57.42 | 64.54 | 85.93 | 95.17 | 89.00 | 83.80 | 93.06 | 87.44 |
| Human Level | 86.68 | 81.00 | 71.00 | 80.30 | 90.30 | 98.00 | 84.00 | 92.40 | 87.10 | 96.00 |
Certifications
Record Filing Status
| Large Model Registration Code | Beijing-MengZiGPT-20231205 |
| Network Information Security Record Number | 110108396388101240013号 |
| Network Information Security Record Number | 110108396388101240021号 |
Application Scenarios
Bulletin Extraction
Fiction Generation
Sentiment Classification
Research Reports Classification
News Digest
Knowledge Map Construction
Q&A System
Image-text Mutual Inspection
Recommendations
Langboat Smart Knowledge Base
Provide intelligent AI search, AI-assisted writing, and other functions to help enterprises rapidly build their own secure and reliable knowledge mid-platform.
Langboat Meeting Assistant Platform
The Langboat Meeting Assistant can efficiently realize the functions of speech to text and multi-dimensional intelligent analysis of conference. Applicable to office meetings, teaching speeches, media interviews and other meeting scenarios，Provide deep analysis and value mining of meeting contents.
Langboat Smart Agent Builder
An enterprise intelligent Agent application building tool platform.
Products
Business Cooperation Email
Address
Floor 16, Fangzheng International Building, No. 52 Beisihuan West Road, Haidian District, Beijing, China.
© 2023, Langboat Co., Limited. All rights reserved.
Large Model Registration Code：Beijing-MengZiGPT-20231205
Business Cooperation:
bd@langboat.com
Address:
Floor 16, Fangzheng International Building, No. 52 Beisihuan West Road, Haidian District, Beijing, China.
Official Accounts:
© 2023, Langboat Co., Limited. All rights reserved.
Large Model Registration Code：Beijing-MengZiGPT-20231205
