DeepSeek-R1

DeepSeek-R1 is a family of open-source large language models (LLMs) developed by the Chinese artificial intelligence company DeepSeek. Released in January 2025, the models are specifically designed to excel at complex reasoning tasks, including mathematics, coding, and logical deduction. The flagship model, DeepSeek-R1, contains 671 billion total parameters with 37 billion active parameters per token, employing a Mixture-of-Experts (MoE) architecture that activates only a subset of its parameters for each input. The model was released alongside a technical report and made available under the MIT License, allowing free distillation and commercialization.

The release of DeepSeek-R1 marked a significant moment in the development of AI reasoning models. It was the first open research to demonstrate that advanced reasoning capabilities in LLMs could be incentivized purely through reinforcement learning (RL), without relying on supervised fine-tuning (SFT) on human-annotated reasoning trajectories. This approach contrasted sharply with prevailing methods that depended heavily on extensive human-labelled data to teach models how to reason step by step.

Architecture

DeepSeek-R1 is built upon the DeepSeek-V3 Base model and utilizes a Mixture-of-Experts (MoE) framework. The architecture incorporates Multi-head Latent Attention (MLA) and a specialized DeepSeekMoE structure. It also includes built-in multi-token prediction (MTP) for speculative decoding, which improves inference efficiency. The model supports a context length of 128,000 tokens, enabling it to process and reason over long documents and complex problem statements.

The MoE design allows DeepSeek-R1 to maintain a large total parameter count while keeping computational costs manageable during inference. Only 37 billion parameters are activated for any given token, which is a fraction of the total 671 billion. This architecture enables the model to deliver high performance on reasoning-intensive tasks without the prohibitive inference costs typically associated with dense models of comparable scale.

Training Methodology

The training pipeline for DeepSeek-R1 represents a departure from conventional LLM development. The researchers first created DeepSeek-R1-Zero, a model trained via large-scale reinforcement learning directly on the base model, bypassing the supervised fine-tuning stage entirely. The reward signal was based solely on the correctness of final answers, without imposing any constraints on the reasoning process itself.

Through this pure RL training, DeepSeek-R1-Zero spontaneously developed sophisticated reasoning behaviors, including self-verification, reflection, and the exploration of alternative solution paths. The model exhibited a tendency to generate longer responses that incorporated verification steps and backtracking when encountering dead ends. This emergent behavior demonstrated that reasoning capabilities could arise from RL without explicit human instruction on how to reason.

However, DeepSeek-R1-Zero suffered from several practical issues, including poor readability, endless repetition, and language mixing, where the model would occasionally switch between English and Chinese within a single chain of thought. To address these problems, the researchers introduced DeepSeek-R1, which incorporates a small amount of cold-start data before RL training. The final pipeline involves two RL stages aimed at discovering improved reasoning patterns and aligning with human preferences, along with two SFT stages that serve as seeds for the model's reasoning and non-reasoning capabilities.

Performance and Benchmarks

DeepSeek-R1 achieves performance comparable to OpenAI's o1 model across a range of reasoning benchmarks. On the AIME 2024 mathematics competition, DeepSeek-R1 attained a pass@1 score of 79.8 percent, while OpenAI's o1-1217 scored 79.2 percent. On MATH-500, the model achieved 97.3 percent pass@1, slightly surpassing o1's 96.4 percent. For coding tasks, DeepSeek-R1 reached a Codeforces rating of 2029, outperforming o1's 1891 rating. On the GPQA Diamond benchmark, which tests graduate-level scientific reasoning, DeepSeek-R1 scored 71.5 percent pass@1 compared to o1's 75.7 percent.

Beyond the flagship model, DeepSeek released six distilled dense models based on the Qwen and Llama architectures. These smaller models, ranging from 1.5 billion to 70 billion parameters, were fine-tuned using reasoning data generated by DeepSeek-R1. The DeepSeek-R1-Distill-Qwen-32B model outperformed OpenAI's o1-mini across various benchmarks, establishing new state-of-the-art results for dense models. The 70-billion-parameter distilled model achieved an AIME 2024 pass@1 score of 70.0 percent, demonstrating that reasoning patterns from large models can be effectively transferred to smaller ones.

Open-Source Release and Distillation

DeepSeek-R1 was released under the MIT License, making it one of the most permissively licensed frontier reasoning models available. The open-source release included the full model weights for both DeepSeek-R1 and DeepSeek-R1-Zero, along with the six distilled models. The technical report was published on arXiv and later in the journal Nature, providing the research community with detailed insights into the training methodology and architecture.

The decision to open-source the model and its distilled variants had significant implications for the AI research community. Researchers could now study and build upon a reasoning model that rivaled proprietary systems, and developers could fine-tune the models for specialized applications. The MIT License explicitly permitted distillation and commercialization, removing legal barriers that often accompany the release of advanced AI models.

Impact and Reception

The release of DeepSeek-R1 generated considerable attention in both the AI research community and the broader technology industry. Its performance on par with OpenAI's o1 model, combined with its open-source nature, challenged assumptions about the necessity of massive proprietary investment for developing frontier reasoning capabilities. The model demonstrated that pure reinforcement learning could incentivize sophisticated reasoning behaviors without the need for extensive human-labelled reasoning data, a finding that has since influenced subsequent research in the field.

The distilled models released alongside DeepSeek-R1 also had a notable impact, particularly for researchers and developers with limited computational resources. The DeepSeek-R1-Distill-Qwen-32B model, for instance, achieved performance exceeding OpenAI's o1-mini on several benchmarks despite being substantially smaller and more efficient to run. This made advanced reasoning capabilities accessible to a wider range of users and applications.

The model's architecture and training methodology have been widely studied and replicated. The use of Group Relative Policy Optimization (GRPO) as the RL framework, which eliminates the need for a separate value network, has been adopted in other reasoning model projects. The concept of using pure RL to incentivize reasoning, bypassing SFT on human demonstrations, has become a prominent research direction.

References

DeepSeek API Docs. "DeepSeek-R1 Release." 

Nature. "DeepSeek-R1 incentivizes reasoning in LLMs through reinforcement learning." 

SGLang Documentation. "DeepSeek-R1." 

Hugging Face. "README.md · qgallouedec/DeepSeek-R1 at main." 

[Read more →]

Google Gemini

Google Gemini is a family of advanced multimodal artificial intelligence (AI) models developed by Google. Introduced as the successor to Google Bard, Gemini is designed to be an AI assistant capable of understanding and processing various types of input, such as text, images, audio, video, and programming code. Since its official launch in February 2024, Gemini has evolved into one of the most comprehensive AI ecosystems, deeply integrated with Google services and available in various subscription tiers to meet the needs of users ranging from students to enterprises.

History and Development

The roots of Google Gemini lie in Google Bard, a chatbot based on the LaMDA language model that was launched in 2023. A major transformation took place in early 2024 when Bard was upgraded with a more advanced next-generation large language model (LLM). On February 8, 2024, Google officially announced the name change from Bard to Gemini, marking a strategic shift to unify all of its AI capabilities under a single brand umbrella. This change was also accompanied by the launch of the Gemini app for smartphones, which gradually replaced Google Assistant as the primary digital assistant on certain Android devices.

Throughout 2025 and 2026, Google continued to develop Gemini models at a high pace. The launch of models such as Gemini 3.5 Flash and Gemini 3.1 Flash-Lite demonstrated Google's commitment to providing models with a balance between high intelligence and low latency, especially for programming and agentic tasks. Recent innovations include the introduction of Gemini Omni Flash, a multimodal model capable of generating 720p-quality video from text descriptions or editing video through simple conversation. This model was released in public preview in late June 2026, marking a major step for Google in the field of generative content creation.

Core Capabilities and Features

Multimodality and Content Processing

Gemini's main feature is its capability as a multimodal model. Users can interact with Gemini through text, voice, and images. The Gemini app allows users to take photos and ask questions about objects around them. This capability is extended with the video-to-image feature on the latest model, in which Gemini can generate high-quality thumbnails or posters from uploaded video files. Gemini is also able to create Audio Overviews, turning documents and reports into engaging podcast-style discussions, making complex content easier to consume.

Integration and Extensions

Gemini is designed to work seamlessly with the Google ecosystem. Through the extensions feature, Gemini can connect with apps such as Gmail, Google Maps, YouTube, and Google Drive to help complete tasks more quickly. This capability allows Gemini, for example, to summarize email contents, search for travel routes, or find information from videos without having to open the apps separately. On Pixel 9 devices and later, Gemini is the default assistant, accessible by long-pressing the power button or using the voice command "Hey Google".

Personalization with Gems

One of the main personalization features is Gems. Gems are specialized versions of Gemini that can be customized to become "experts" on a particular topic without requiring coding skills. Users can create Gems for specific purposes such as writing editors, event planners, or sentiment analysis, by uploading instructions and reference files. Google also provides ready-made Gems, such as Learning coach for study guidance and Career guide for career preparation, which are highly useful in the education sector.

Gemini Live and Conversation

The Gemini Live feature delivers a more natural conversational experience. Users can speak with Gemini in real time, use it to brainstorm ideas, or simulate important conversations. Gemini Live is designed for smooth interaction, allowing users to share their screen or camera while speaking in order to receive contextual responses.

Models and Versions

Google releases Gemini in various model variants to meet different needs and budgets. Each model has optimized specifications, ranging from speed and cost to intelligence and multimodal capability.

Gemini 3.5 Flash

This model is a mainstay for tasks that require substantial intelligence at high speed. Released generally in May 2026, Gemini 3.5 Flash is known to be highly effective for programming tasks and long-term "agentic" tasks, offering performance comparable to large "flagship" models at a lower cost.

Gemini 3.1 Flash-Lite

This variant is optimized for speed, scale, and cost efficiency. The model is well suited to simple tasks and image generation with ultra-low latency.

Gemini Omni Flash

The newest model, still in public preview. Its main capability is conversational video creation and editing. Equipped with an invisible SynthID digital watermark for content verification, this model represents a new generation of generative AI capable of creating content from various types of input.

Image Models (Nano Banana)

Gemini also has specialized visual models such as Gemini 3.1 Flash Image ("Nano Banana 2") and Gemini 3 Pro Image ("Nano Banana Pro"). These models have been generally available and excel at image creation and editing, including a feature for generating images from video.

Availability and Subscriptions

Gemini can be accessed through various platforms, including mobile apps for Android (Android 10+) and iOS, as well as through the website at gemini.google.com. Google offers several subscription plans.

The Google AI Pro plan unlocks access to the most advanced models such as 2.5 Pro, the Deep Research feature for detailed reports, and eight-second video generation with Veo 3, as well as a 1-million-token context window capable of processing up to 1,500 pages of text.

The Google AI Ultra plan is the highest tier, offering exclusive access to the most powerful models, the latest features such as Agent Mode, and the opportunity to try Google's AI innovations earlier. This plan is available in certain countries and is not for Workspace customers.

In Southeast Asia, Gemini has recorded significant user growth, more than doubling in a year. Vietnam has become the regional leader in using Gemini for academic purposes and local-language use. The country has even become the market with the highest percentage of native-language (Vietnamese) use in the region, reaching 89%.

Impact and Education

Gemini, along with other generative AI, has begun to change the way people access information. This change has sparked discussion about the future of traditional knowledge sources such as Wikipedia. Wikimedia Indonesia, for example, held the #WikipediaxAI program in 2025 to discuss how AI uses Wikipedia content and how the community can adapt. Concerns have arisen that the shift from reading articles to receiving instant summaries from AI could reduce the number of visitors to source sites and the number of new volunteer contributors.

On the other hand, Gemini offers great potential in education. In Vietnam, more than 160,000 students use the Gemini Canvas feature every month for exam preparation, while educators generate tens of thousands of teaching assistance requests every day. This use shows AI's role as a powerful supporting tool, not a replacement for human critical thinking. A comparative study even placed Gemini as a tool comparable to ChatGPT for various translation tasks, demonstrating its capability in language processing.

References

1. Use Gemini, your AI assistant, on your Pixel phone. Google Help.

2. Education/News/December 2025/WikipediaxAI: Wikipedia, AI, and the future of knowledge. Meta-Wiki, Wikimedia.

3. Catatan rilis, Gemini API. Google AI for Developers.

4. Personalize Gemini with Gems: custom AI experts for any topic. Google Services.

5. Google Gemini, Apps on Google Play.

6. Cara Menggunakan Gemini AI Google Bahasa Indonesia, Mudah dan Praktis. Kompas.com.

7. Release notes, Gemini API. Google AI for Developers.

8. Gemini, "Asisten" Canggih untuk Sehari-hari. LinkedIn, 360 Solusi Teknologi.

9. Việt Nam dẫn đầu Đông Nam Á về dùng AI của Google hỗ trợ học thuật. VietnamPlus.

10. Digilib UIN SGD (Bab 1).

11. 100 things we announced at I/O 2026. Google Blog.

12. Try Gemini, your personal AI assistant. Android.com.

[Read more →]

ChatGPT

ChatGPT is an artificial intelligence (AI) model developed by OpenAI, designed to generate text similar to human writing based on input provided by the user. Launched in November 2022, ChatGPT quickly became a global phenomenon and changed the way society interacts with technology, from professional work to everyday personal use. Its presence has not only revolutionized productivity but also sparked in-depth discussion about the future of knowledge, education, and social interaction in the digital era.

History and Development

The technology underlying ChatGPT is OpenAI's GPT (Generative Pre-trained Transformer) model series, which has undergone a number of significant improvements. Since its debut with GPT-3.5, the model has transformed through various versions up to GPT-5.5, which has become the most frequently used model. Each new generation has brought improvements in reasoning ability, following of complex instructions, and overall conversation quality.

This development marks a shift from merely a language model to a more intelligent assistant. In mid-2026, models such as GPT-4.5 and GPT-5.2 were officially retired in ChatGPT and shifted to newer versions to ensure users receive the best experience. In addition, OpenAI periodically releases new models such as GPT-5.5 Instant Mini, which functions as a backup model when users reach usage rate limits, ensuring continued access to the service.

Main Features and Capabilities

ChatGPT has evolved from merely a chat box into an advanced platform with a variety of features designed to increase productivity and personalization.

Context and Personalization

One of ChatGPT's main capabilities is memory and personalization. This feature allows ChatGPT to remember information from previous conversations, providing more relevant and contextual responses over time. Users can view a memory summary, correct stored information, or turn off the memory function entirely to maintain privacy. This personalization makes ChatGPT an assistant that is increasingly adaptive to each user's specific needs.

Scheduled Tasks

The "Scheduled Tasks" feature allows users to ask ChatGPT to perform repeating work, send reminders, or monitor something at a predetermined time. This changes ChatGPT from a reactive tool into a proactive tool that can be relied on for various daily routine needs.

Codex and Work Automation

The development of Codex marked a major leap in ChatGPT's operational capabilities. Features such as Codex Remote allow users to control a remote computer from a phone, while Record & Replay allows users to demonstrate a workflow and turn it into a reusable skill. For business users, this opens large opportunities to automate repeating tasks in various fields, such as software development, data analysis, and project management. The Developer mode for Browser use feature also gives developers more control for debugging and performance analysis.

Productivity and Accessibility Improvements

In addition to the features above, ChatGPT has consistently introduced improvements to the user experience. Improvements to the dictation model with a word error rate lower by up to 10% make it easier for users to interact without typing. For users who enter long text, ChatGPT now automatically converts pastes of more than 10,000 characters into attachments, keeping the chat box tidy. Pronunciation assistance features for more than 60 languages and the integration of current information updates such as World Cup schedules also add to ChatGPT's usefulness.

Impact on Knowledge and Education

The presence of ChatGPT has had a major impact on the way humans access and process information. As a language model trained on large amounts of data from the internet, including Wikipedia articles, ChatGPT is able to provide instant and concise answers to user questions. This has changed information-seeking behavior, in which users can obtain knowledge without having to visit the source site directly, such as Wikipedia, which is one of its main training data sources. A study even showed that Wikipedia is the source most frequently cited by ChatGPT, with a share reaching 16.3% of all citations.

Role in Education

In the world of education, empirical studies have explored ChatGPT's potential as a cognitive scaffold. Research in 2026 at a madrasah in Indonesia showed that the use of ChatGPT in argumentative writing instruction had a large effect on student achievement, particularly in the aspects of organization and written content. ChatGPT was shown to help students construct a framework of thinking and develop arguments, although its effectiveness was not even across all linguistic aspects such as vocabulary and grammar. Another study also looked at the use of ChatGPT in ecology-based Indonesian language learning, opening opportunities for more innovative teaching methods.

Knowledge Dynamics in the AI Era

Experts see that generative AI, including ChatGPT, will change the knowledge landscape. There are concerns that if AI becomes increasingly advanced, the role of encyclopedias such as Wikipedia as a center of knowledge could be displaced. However, there is also a view that AI can instead become a tool for democratizing knowledge, by translating and disseminating information into various regional languages that were previously difficult to access. This discussion highlights the importance of the role of human curators in ensuring the quality and validity of information, even if that content is generated by AI.

Regulation and Governance

Along with its growing popularity, ChatGPT and other digital platforms have faced oversight from regulators in various countries. In Indonesia, in November 2025, the Ministry of Communication and Digital Affairs (Komdigi) issued a warning to dozens of major digital platforms, including OpenAI (ChatGPT) and Wikipedia, because they had not yet registered as Private-Scope Electronic System Operators (PSE). This step was taken to ensure compliance with Minister of Communication and Informatics Regulation Number 5 of 2020, with the threat of sanctions in the form of access blocking if not fulfilled immediately.

Social Phenomenon: Confiding in AI

The phenomenon of ChatGPT use is not limited to a productivity tool, but has also entered the social-emotional realm. Digital sociology research in late 2025 revealed that many users in Indonesia use ChatGPT as a "confidant." The main factors behind this phenomenon are the perception that AI is non-judgmental, always available 24 hours a day, provides emotional validation, and offers a sense of safety in terms of privacy. However, this research also warned of potential negative impacts, such as emotional dependence, widening social distance, and the risk of data leaks. This highlights the need for balanced digital literacy so that AI does not replace social relations among humans.

References

· Wikimedia. (2025, Desember 15). Education/News/December 2025/WikipediaxAI: Wikipedia, AI, and the future of knowledge. Meta-Wiki. 

· OpenAI Help Center. (2026, Juni 26). ChatGPT — 版本说明. 

· OpenAI Help Center. (2026, Juli 6). ChatGPT — Release Notes. 

· OpenAI Help Center. (2026, Juni 26). ChatGPT Business - 发行说明. 

· OpenAI Help Center. (2026, Juni 26). ChatGPT Business - 發布說明. 

· Hops.ID. (2025, November 19). Cek HP Anda Sekarang! 25 Aplikasi Favorit Ini Terancam Hilang, Ada ChatGPT hingga Wikipedia. 

· Bennylin. (2023, Mei). Pengguna:Bennylin/20TahunWikipedia. Wikipedia bahasa Indonesia. 

· Syafawani, R., dkk. (2025, Desember 23). Fenomena Curhat ke AI: Tinjauan Sosiologi Digital Terhadap Pergeseran Interaksi Sosial di Indonesia. Jurnal Sosial dan Sains. 

· Jumaddil, J., dkk. (2026, April 30). ChatGPT as a Macro-Level Cognitive Scaffold for First-Language Argumentative Writing. Edcomtech. 

· Steavenson, M. (2025, Juni 14). AI Search Visibility: Why Reddit, Wikipedia, and YouTube Win And What It Means For Your Brand. LinkedIn. 

· Abdian Amien, A., & Kusumawati, H. (2024, Desember). Optimalisasi CHATGPT Dalam Pembelajaran Bahasa Indonesia Berbasis Ekologi Di SMAN 4 Pamekasan. GHANCARAN. 

[Read more →]