Alibaba Unveils Qwen3.8-Max AI Model with 2.4 Trillion Parameters, Escalating U.S.-China AI Race

China's Alibaba takes another swipe at America's AI supremacy

Chinese tech giant Alibaba released what it says is its largest and "most capable AI model to date," claiming performance rivalling the best systems from US frontier labs Anthropic and OpenAI, as well as domestic rivals like Moonshot AI's Kimi K3. Alibaba said it was making the model, Qwen3.8-Max, widely available to users in a blog post published on Monday. The release had been expected after the company previewed the model last month, when it claimed it was "second only to Fable 5," Anthropic's flagship. The open release of another highly capable Chinese AI model adds to sky-high tensions along multiple fronts in Silicon Valley and Washington over how to safely manage AI systems and retain the US' technological edge over China. Results from Alibaba's own testing shared on Monday suggests it's as powerful as claimed, as does its ranking on crowdsourced model-comparison platform Arena.AI. Alibaba's own testing shows the model's performance to broadly match -- and sometimes exceed -- that of Fable 5 on benchmark tests. On the Arena text model leaderboard, Qwen3.8-Max trails only Fable 5 and three models in Anthropic's Opus family. For frontend coding, it is beaten only by two Claude Opus models and Kimi K3, and for visual analysis, only Fable 5 trounces it. Alibaba says Qwen3.8-Max has 2.4 trillion parameters, a numerical measure of the settings a model learns during training that it uses to process data, recognize patterns, and undertake various tasks. While parameter counts are widely used as a shorthand for model performance, a bigger number does not always mean better. Moonshot's Kimi K3 model has 2.8 trillion parameters, but most top American labs keep figures private and neither OpenAI nor Anthropic disclose exact counts for their top systems. The Chinese company said it will release the weights for Qwen3.8-Max next week. Weights are the adjustable numerical values of an AI that determine how it processes information. Open-weight systems, while more restrictive than traditional open-source software, give developers far more control than they have over proprietary products from companies like OpenAI and Anthropic. Qwen3.8-Max marks a return to open-weight releases for Alibaba after the company briefly pivoted towards proprietary releases for its more advanced models earlier this year. Open-weight releases have become a growing point of differentiation for China's AI industry, where they have become the norm. Moonshot released Kimi K3's last week and many other top AI models are also open-weight. Beijing has championed the strategy as a means of growing China's influence in global AI governance and encouraging widespread adoption of domestic tech champions. Alibaba's release intensifies competition with China, whose firms appear to be rapidly narrowing the gap with US companies and have increased the tempo of releases in recent weeks. Qwen3.8-Max closely follows the release of Kimi K3, viewed as another challenge to American AI dominance, and both ByteDance and MiniMax released capable new video generation models on Friday. Openness has also proved to be a divisive issue. Amid reports of potential crackdown on open tools in the wake of the Chinese releases, the US industry has largely rallied around preserving access to open-weight models, both as a safety necessity and as means of preserving competition. That debate comes as closed-model providers, notably OpenAI and Anthropic, face increasing scrutiny after revealing a slew of cyberattacks unknowingly perpetrated by their own escaped AI agents. Incident reports from one victim suggest the restrictive safety rails intended to stop nefarious use of AI models also limits their use as defensive tools.

Alibaba takes aim at OpenAI and Anthropic with Qwen3.8-Max launch

The new open-weight model targets enterprise software engineering and multimodal workloads as enterprises weigh deployment efficiency alongside AI model performance. Alibaba on Monday introduced Qwen3.8-Max, its largest artificial intelligence model to date, expanding its enterprise AI portfolio with an open-weight model designed for software engineering, multimodal reasoning, and other knowledge-intensive business workloads. In a blog post announcing the launch, Alibaba described Qwen3.8-Max as a 2.4-trillion-parameter mixture-of-experts (MoE) model that activates only about 95 billion parameters during inference. The company said the architecture is intended to improve inference efficiency while supporting coding, reasoning and multimodal tasks, with open-weight versions scheduled for release next week through Alibaba Cloud's Model Studio.

DeepSeek's new AI model is by far the cheapest of well-known models to run, research firm says

BEIJING, Aug 3 (Reuters) - A version of Chinese startup DeepSeek's flagship AI model is by far the least expensive to run on benchmark tests among well-known models globally and more than 100 times cheaper to run than Anthropic's Claude Fable 5, according to a research firm. DeepSeek, which sources have said is preparing for a potential IPO, officially released its V4-Flash model on Friday, its latest attempt to regain momentum by doing what it is best known for - offering ultra-low-cost AI alternatives. The startup's R1 model became a global sensation in early 2025, triggering ⁠a selloff in global technology stocks and raising questions about the large amounts U.S. companies were spending on AI. DeepSeek's V4-Flash charges $0.14 per million input tokens and $0.28 per million output tokens, according to research firm Artificial Analysis. A token is a unit of data used to measure AI usage. San Francisco-based Artificial Analysis estimated V4-Flash's average cost at 3 cents per test, compared with 86 cents for Kimi K3 from Chinese rival Moonshot AI, $1.86 for OpenAI's GPT-5.6 Sol and $3.15 for Claude Fable 5. The comparison provides a more realistic measure of value than pricing alone because it accounts for the amount of data a model must process and generate to complete a task. A model with low headline price can still prove expensive if it ⁠requires significantly more steps to produce an answer. DeepSeek once commanded most of the headlines about Chinese AI development but was quickly besieged by many domestic rivals including other startups such as Moonshot, MiniMax and Z.AI as well as tech giants like ByteDance and Alibaba (9988.HK), opens new tab. All are vying with U.S. tech firms for global adoption, targeting businesses seeking cheaper ways to deploy AI at scale. Artificial Analysis said DeepSeek's ⁠V4-Flash model scored 50 out of 100 on its Intelligence Index, which combines results from nine benchmarks spanning coding, reasoning and workplace-style assignments. That's the same score as Google's (GOOGL.O), opens new tab Gemini 3.6 Flash, and one point behind Meta's (META.O), opens new tab Muse Spark 1.1 and GLM-5.2 from Z.AI which ⁠is also known as Zhipu. Moonshot's Kimi K3, however, scored a 57 while Anthropic's Claude Opus 5, Fable 5, and OpenAI GPT-5.6 scored nine or more points higher. DeepSeek is also preparing a more powerful version of its model, called the ⁠V4-Pro. It has not given a date for that version's official release. Separately on Monday, Alibaba unveiled its largest and most capable artificial-intelligence model to date, the Qwen3.8-Max, which is not far behind in size when compared with an offering from domestic rival Moonshot AI launched last month. Reporting by Eduardo Baptista; Editing by Miyoung Kim and Edwina Gibbs Our Standards: The Thomson Reuters Trust Principles., opens new tab * Suggested Topics: * Artificial Intelligence Eduardo Baptista Thomson Reuters Eduardo Baptista is a Senior Correspondent for Reuters based in Beijing, covering China's technology, space, and automotive industries. He has led enterprise and investigative reporting on China's military-linked companies, artificial intelligence and semiconductor supply chains, as well as macroeconomic and industrial policy. Baptista has reported from China for nearly a decade and holds a BA in History from the University of Cambridge.

Alibaba shares rally after unveiling its 'most powerful' AI model as U.S.-China competition heats up

Alibaba unveiled its latest and "most powerful" AI model Qwen3.8-Max on Monday as Chinese companies race to close the AI gap with the U.S. Qwen3.8-Max, which is scheduled for release next week, boasts 2.4 trillion parameters, making it one of the "most powerful" AI models in the Chinese tech giant's Qwen family of AI models to date, Alibaba said. Parameters refer to the numerical settings that shape how AI processes information and generates responses. The model also supports a context window of up to 1 million tokens, which means it can understand and work with thousands of pages of information. Alibaba's New York-listed shares were up 4.5% in premarket trading, while its shares rose 7% on the Hong Kong exchange.

Alibaba unveils Qwen3.8-Max, its most capable model, closing on Moonshot in size

The 2.4-trillion-parameter model ranks top among Chinese text models and second worldwide on a visual benchmark, and launches next week. Alibaba has unveiled Qwen3.8-Max, the most capable model it has built and a pointed entry in the size race between China's largest AI developers. At 2.4 trillion parameters it sits just below Moonshot's Kimi K3, which carries 2.8 trillion, close enough that the comparison is the story. The model is multimodal, handling text, images, and video, and can take up to a million tokens in a single prompt. Alibaba says it will launch next week through Model Studio, the developer platform on Alibaba Cloud. On Arena.AI's public leaderboard, Qwen3.8-Max ranks highest of any Chinese text model and second in the world on the visual-analysis benchmark, behind only Anthropic's Claude Fable 5. It is a step up from the version Alibaba recently billed as the world's No.2 AI model. The headline parameter count is not the number that governs cost. Qwen3.8-Max uses a mixture-of-experts design that activates only about 95 billion parameters for any given request, a way of keeping a very large model cheap to run and quick to answer. A million-token context window matters for the tasks Alibaba is chasing. It is enough to hold a large codebase, a long video transcript, or a stack of documents at once, the raw material for the agent work the company keeps circling back to. That Alibaba is quoting parameter counts at all marks a divide in the industry. Chinese developers have made openness a selling point, publishing sizes and often weights, while OpenAI, Anthropic, and Google keep those figures to themselves. Alibaba also says the model completed a software-engineering project over a 16-day autonomous run, a claim that points at the growing interest in long-horizon agent tasks. The figure is the company's own and has not been independently tested. The release keeps Alibaba in close contact with Moonshot, whose Kimi K3 has been the model to beat this year. Demand ran hot enough that Moonshot paused new sign-ups to protect capacity, a sign of how fast a strong Chinese model now finds users. Size is only part of the push. Alibaba has been building out AI models for robots as China's attention shifts from chatbots to agents that can act, and Qwen is meant to be the brain those products call on. There is a commercial engine underneath all of it. Alibaba has been folding Qwen into its own services, from cloud to consumer shopping, which gives each new model an immediate route to real users rather than a standalone demo. The launch continues a fast release cadence from the Qwen team, which has shipped a steady run of models this year. The tempo is part of the strategy, keeping Alibaba in the conversation each time a rival claims the lead. Giving models away, or close to it, is a deliberate wedge. Open weights win developer mindshare and pull workloads onto Alibaba Cloud, where the company can still charge for the compute that runs them, whatever the model itself costs. None of it comes with a price yet. Alibaba did not publish token pricing for Qwen3.8-Max, though its recent models have undercut Western rivals sharply, and the wider Chinese market has been racing costs toward the floor. The rise has not been frictionless. Anthropic has accused Alibaba of running its largest distillation campaign against Claude, alleging attempts to copy the American model's behaviour, a charge Alibaba disputes. For buyers outside China, the question is less about the top of a leaderboard than whether an open, cheaper model is close enough to the frontier to switch to. On these benchmarks, Alibaba is arguing that it is. Whether Qwen3.8-Max holds those rankings once developers get their hands on it next week is the open question. Leaderboards move quickly, and in Chinese AI right now they move faster than most.

Alibaba unveils its most capable AI model to date, not far behind Moonshot's in size

BEIJING, Aug 3 (Reuters) - China's Alibaba (9988.HK), opens new tab on Monday unveiled what it said is its largest and most capable artificial-intelligence model, the Qwen3.8-Max, which is not far behind in size when compared with an offering from domestic rival Moonshot AI launched last month. Chinese tech companies -- a huge force in open-weight AI models globally -- are locked in a fierce and fast-moving battle to build more powerful systems without making them prohibitively expensive to run. Qwen3.8-Max has 2.4 trillion parameters, the numerical settings a model learns from data ⁠and uses to recognise patterns, generate answers, and carry out tasks. Moonshot's Kimi K3 has 2.8 trillion parameters. A higher figure does not automatically make a model better, but it has become a closely watched measure of the scale of the computing and data behind advanced AI systems. Chinese tech companies are keen to publish parameter count to help their models gain traction among the developer community. Their models tend to be open-weight, meaning the underlying learned settings that allow developers to run or adapt the system are available for download. By contrast, OpenAI, Anthropic and Google (GOOGL.O), opens new tab do not publish parameter count for their closed-source models. Qwen3.8-Max was unveiled on crowdsourced, model-comparison platform Arena.AI, ⁠where it immediately became the highest-ranking Chinese model in terms of text models, though it still lags Claude Fable 5 and three Opus variants which are all from Anthropic. But on Arena.AI's leaderboard for AI models that analyse images and other visual material, Qwen3.8-Max ranked second globally, only behind a Claude Fable 5 variant. Both Qwen3.8-Max and Kimi K3 can handle text, images ⁠and video, and process up to 1 million tokens at a time. Tokens are chunks of data, often parts of words or short words, and a big figure means the model can take in large amounts of material in one go, such ⁠as long legal files, a large software codebase or hundreds of pages of documents. Alibaba said its model uses a "mixture-of-experts" design, which divides work among specialised parts of the system instead of switching on the entire ⁠model for every request. Only 95 billion parameters are used at a time, reducing costs and response delays. The tech giant said the model completed a software-engineering project in 16 days. The Qwen3.8-Max is due to be released next week through Alibaba Cloud's Model Studio platform. Reporting by Eduardo Baptista; Editing by Edwina Gibbs Our Standards: The Thomson Reuters Trust Principles., opens new tab * Suggested Topics: * Retail & Consumer Eduardo Baptista Thomson Reuters Eduardo Baptista is a Senior Correspondent for Reuters based in Beijing, covering China's technology, space, and automotive industries. He has led enterprise and investigative reporting on China's military-linked companies, artificial intelligence and semiconductor supply chains, as well as macroeconomic and industrial policy. Baptista has reported from China for nearly a decade and holds a BA in History from the University of Cambridge.

DeepSeek's V4-Flash is the cheapest well-known AI model to run, research firm finds

Artificial Analysis put the cost of running the model through its benchmark suite at about three cents, a fraction of what rivals charge. It now costs about three cents to push one of the world's better-known AI models through a full benchmark suite. That is the figure from Artificial Analysis, which found that a version of DeepSeek's flagship model, V4-Flash, is by far the cheapest well-known model to run. The independent research firm measured the cost of completing its Intelligence Index test battery on each model. V4-Flash came in at roughly three cents, and the nearest comparisons were not close: Moonshot's Kimi K3 cost 86 cents, OpenAI's GPT-5.6 Sol $1.86, and Anthropic's Claude Fable 5 $3.15. On published pricing, DeepSeek charges $0.14 per million input tokens and $0.28 per million output tokens for the model. Those are the sort of numbers that quietly redefine what 'expensive' means at the frontier. V4-Flash is the lighter half of the pair DeepSeek shipped when it returned with V4-Pro and V4-Flash, with the heavier V4-Pro aimed at harder reasoning. Flash is built for volume: fast, cheap, and good enough for a large share of everyday work. The catch is capability, and the numbers are honest about it. V4-Flash scored 50 out of 100 on the Intelligence Index, level with Google's Gemini 3.6 Flash and just behind Meta's Muse Spark 1.1 and Z.ai's GLM-5.2, both on 51. The frontier still sits clearly ahead. Kimi K3 scored 57, while Claude Opus 5, Claude Fable 5, and GPT-5.6 landed roughly nine points higher again, a reminder that cheapest and best remain different questions. The release lands in the middle of an AI price war that DeepSeek has done more than anyone to start. The company made a 75% discount permanent earlier this year, and rivals have been cutting in response. Those rivals have been moving the same way. OpenAI trimmed GPT-5.6 pricing sharply, and the general drift of the market has been down, and fast, on a curve that looks less like software margins and more like a commodity. DeepSeek has the balance sheet to keep pushing. The company recently closed its first outside funding, a round of more than $7bn, which buys room to subsidise aggressive pricing while it takes share. Flash-class models are aimed at the high-volume end of the market. That means the chatbots, coding assistants, and back-office automation where requests run into the millions and every fraction of a cent compounds into a real bill. DeepSeek's edge is as much engineering as pricing. The company has leaned on efficient training and inference to hold costs down, which is what lets it charge so little without, it says, simply setting money on fire. That trend carries consequences beyond a cheaper API bill. Analysts have argued that relentless discounting from Chinese labs puts the eventual OpenAI and Anthropic IPOs under pressure, since premium pricing is hard to defend when a rival is tens of times cheaper per task. Not everyone is convinced the quality gap still matters. Zack Kass, OpenAI's former head of go-to-market, has framed the moment as one of 'diminishing model returns', arguing that once models are close enough, the next one barely moves the needle and price does the deciding. Chinese labs have been setting that pace. Moonshot's Kimi K3 spooked markets on release, and the broader worry is that a wave of cheap, open-weight models erodes the economics Western AI valuations quietly assume. Benchmarks are an imperfect proxy, and cost per test turns on how efficiently a model spends tokens as much as on its sticker price. Even so, the direction is not in doubt, and Artificial Analysis has put hard figures on what developers have felt for months. For buyers, the sum is getting simpler. If a model that costs three cents to run can do most of the job, the burden shifts onto the expensive models to prove what those extra nine points on a benchmark are really worth.

Alibaba launches Qwen3.8-Max, its largest AI model yet

On coding, Alibaba said the model ran for more than 16 days in a fully autonomous operation, accumulating 265 commits, 127 pull requests, and 151 issues in a self-built software repository without human input. In a separate test, it entered a competition against 526 human teams, completing tasks under a 24-hour limit and finishing ahead of 87% of the field. For document and video work, the model can ingest financial reports exceeding 200 pages and footage running beyond 100 hours, then compile that material into structured knowledge bases that users can search, the company said. It can also rebuild software applications by reading screenshots and turn two-dimensional architectural floor plans into three-dimensional renderings.

Alibaba Just Gave Away Its Best AI Model For Free, Almost Matching Claude and ChatGPT

Alibaba released Qwen3.8-Max on Monday, calling it the most capable model it has ever built. The weights land on Hugging Face and ModelScope next week -- the first time the company has given away a model at Max scale. The specs are big: 2.4 trillion parameters total, with 95 billion switched on at any moment. Parameters are the number of dials a model is able to handle. This is extremely important for efficiency as it means it would not require too much resources to run. Think of it as a huge library where only the relevant shelf lights up for each question. Small businesses and labs with good enough hardware are now able to run a state-of-the-art model without spending the same as a huge datacenter. Alibaba's release post skips the usual benchmark-chasing narrative and leans on endurance instead. The model spent 16 days building a coding tool by itself -- 265 commits, 127 pull requests, 151 issues, no human touching the keyboard. It spent five days reproducing a research paper it had never seen code for, then beat the paper's own results by 2.7 points. In a 24-hour machine learning contest, it finished ahead of 458 of 526 human teams. Built to run inside a rival's tools Qwen3.8-Max ships with instructions for Claude Code and Codex, the coding tools made by Anthropic and OpenAI. Alibaba's API speaks both companies' protocols. Most of its coding benchmarks were run inside Claude Code. And those benchmarks don't flatter it. Across 31 text tests, Anthropic's Fable 5 takes 15 first place spots, OpenAI's GPT-5.6 Sol takes nine, Qwen takes seven. On the 12 coding tests, Qwen wins exactly one. However, in terms of intelligence costs, this model is extremely cheap and efficient, which means that even if it requires more iterations or reasoning, the cost of getting the job done will be much lower, nearly 30% of what Claude Fable 5 charges. That said, flip to multimodal work -- documents, video, spatial reasoning -- and the rankings invert. Qwen leads most of that table. There is also a reversal in terms of business strategy. In April, Alibaba killed the free tier of Qwen Code, with the team drifting toward closed, paid models after leadership departures. Our review of Qwen 3.7 Max noted the Plus version would be open while Max stayed locked behind the API. That door is now open, and the timing isn't an accident. Chinese open-weight models went from under 2% of tokens on OpenRouter in late 2024 to roughly 61% by mid-2026. Qwen passed Meta's Llama as the most self-hosted model in the world. Meanwhile Washington restricted Fable 5 and Mythos 5 under export controls in June, and Beijing is reportedly weighing limits of its own on Chinese models going overseas. So Alibaba is losing on paper and winning on distribution. If you can download something that comes close for free, second place is a fine place to be.

Alibaba's new AI claims to match Claude, upping the US-China AI race

Alibaba's Qwen team says its newest model can design a computer chip and rewrite a research paper -- all without a human watching over it, matching skills its rival Anthropic already claims for Claude. Chinese tech giant Alibaba says its newest AI model can work alone, unsupervised, for days on end and still deliver results, a feat claimed almost exclusively, until now, by its American rival Anthropic. The model, Qwen3.8-Max, was announced in a post by the Qwen team on Monday, which says it delivers "comprehensive improvements across coding, work, research, and long-horizon tasks". The claim echoes how Anthropic's own top model, Claude Fable 5, has been described in the coverage of its release. According to the company, one test saw the model work "completely on its own for about five days or 125 hours of continuous effort," rebuilding a maths research paper's experiment from scratch and then improving on it. In a separate test, Qwen said the model was entered into a live online contest against 526 human teams and, working to a 24-hour deadline, beat all but 68 of them. Similar claims from a US rival Anthropic is making similar claims about its own AI. Its Claude Code tool can read software, plan out fixes and test its own work with little human input. The company's newest model, Claude Opus 5, launched on 24 July, is built to run unsupervised for long stretches and Anthropic says it handles unfamiliar problems far better than earlier versions. Anthropic also sells Claude Cowork, designed for people to hand off longer jobs, such as research, spreadsheets and first drafts, for the AI to finish largely on its own. Neither company's figures have been verified by independent researchers, and both are presenting results that reflect well on their own products. Why it matters The release is the clearest sign yet that China and the United States are now racing each other on the same track. Washington has spent the past two years restricting the export of advanced chips to Chinese AI firms, aiming to keep companies such as Alibaba a generation behind. Alibaba's answer, in effect, is to show it is not behind at all and to go further than its US rivals by making Qwen3.8-Max's weights freely available, inviting the world to check its claims rather than take them on trust. Investors and governments alike are watching for signs of how far apart Chinese and American AI development actually remains.

Alibaba unveils open-source Qwen3.8-Max AI model

Alibaba on Monday unveiled Qwen3.8-Max, its largest open-source artificial intelligence model, and said it would release the model's weights for public download next week. The model has 2.4 trillion parameters and Alibaba said its performance is competitive with leading models from OpenAI and Anthropic. Hong Kong-listed Alibaba shares jumped sharply in early trading after the announcement. Alibaba said the release marks a return to its open-source strategy after the company kept several recent flagship releases proprietary earlier this year. The company said Qwen3.8-Max will be the first Max-class Qwen model to be open-sourced. Qwen3.8-Max supports a context window of up to 1 million tokens and uses 95 billion active parameters out of its 2.4 trillion total, according to Alibaba. The model is multimodal and can process lengthy documents, video content and live streams to build searchable knowledge bases, Alibaba said. Alibaba said the model can recreate software applications from screenshots, generate interactive games and educational animations, and convert two-dimensional floor plans into 3D visualizations. The company said Qwen3.8-Max is designed for autonomous coding, complex research and other long-running agentic tasks. Benchmark tests showed the model was broadly competitive with leading U.S. models, outperforming them on several coding, multimodal and engineering benchmarks while trailing on some general-purpose reasoning tests. Bloomberg reported the model ranks higher on some benchmarks than Moonshot's recently unveiled Kimi K3. The launch comes as competition intensifies among Chinese AI developers including DeepSeek, Moonshot and ByteDance. Earlier this year, DeepSeek's low-cost reasoning models reshaped the market and prompted rapid model upgrades across the industry, according to the source material. Alibaba said in May it would exceed its planned AI infrastructure spending of up to 380 billion yuan, or $55.96 billion, over three years. Qwen3.8-Max is available globally through Alibaba Cloud's Model Studio APIs and through QwenWork, the company's workplace AI agent platform.

Qwen 3.8-Max: can Alibaba challenge OpenAI & Anthropic? | TechPulse

Alibaba has unveiled Qwen 3.8-Max, its largest AI model ever with 2.4 trillion parameters. But what do parameters actually mean, and does a bigger model automatically make it smarter?In this 60-second breakdown, we explain:What AI parameters areWhat tokens are and why they matterQwen's claimed multimodal capabilitiesAlibaba's benchmark results against OpenAI and AnthropicWhy those benchmark scores should be treated as vendor claims until independently verifiedHow Qwen's aggressive pricing ($2 per million input tokens and $6 per million output tokens) could make it a compelling option for startups and enterprisesThe biggest story may not be the model itself, but the pricing strategy. If the performance holds up in independent testing, it could put pressure on AI providers across the industry.

Qwen 3.8 Max Launches with a Massive 2.44 Trillion Parameters

Qwen 3.8 Max is the latest and largest open-weight multimodal AI model, featuring an impressive 2.44 trillion parameters. This advancement marks a new chapter in the Qwen series, emphasizing both scale and functionality. According to Prompt Engineering, one notable feature is Recursive Self-Improvement (RSI), which enables the model to optimize its own processes over time. For example, during a 16-day refinement cycle, the model demonstrated its ability to enhance operational frameworks, showcasing its potential for sustained adaptability. Its open-weight design ensures transparency and accessibility, making it a valuable asset for developers and researchers alike. Discover how Qwen 3.8 Max performs in areas such as Python execution, structured reasoning and creative tasks like animation design. Gain insight into its benchmark results, including its ranking in coding and agentic evaluations. This guide also explores real-world applications, from space tracking to interactive mapping and provides a preview of the upcoming Qwen 3.8 27B, designed to extend advanced AI capabilities to consumer devices. What Makes Qwen 3.8 Max Stand Out? Qwen 3.8 Max distinguishes itself through its remarkable specifications and innovative design. Here are the key features that set it apart: * Unmatched Scale: With 2.44 trillion parameters, including 95 billion actively engaged during typical operations, Qwen 3.8 Max is the first multimodal model in the Qwen series to exceed the trillion-parameter threshold. This scale enables it to process and analyze vast amounts of data with exceptional precision. * Open-Weight Design: The model's open-weight architecture ensures transparency and accessibility, making it a valuable resource for developers and researchers. Platforms like Hugging Face will soon host Qwen 3.8 Max, further expanding its reach. * Scalability and Adaptability: Its versatile architecture is designed to support a broad spectrum of applications, from enterprise-level solutions to creative endeavors, making sure it meets the diverse needs of its users. Advanced Capabilities Qwen 3.8 Max excels in managing long-horizon tasks, making it ideal for handling complex, multi-step processes. One of its most innovative features is Recursive Self-Improvement (RSI), which allows the model to autonomously refine its operational frameworks over time. In a recent demonstration, Qwen 3.8 Max optimized its harnesses over a 16-day period, showcasing its ability to adapt and enhance its performance independently. The model also supports OpenAI-compatible API endpoints, allowing seamless integration into existing workflows. Its advanced capabilities extend to: * Coding and Development: Qwen 3.8 Max can execute Python code, perform structured reasoning and use tools to complete intricate tasks, making it an invaluable asset for developers. * Multimodal Applications: From front-end design to generating animations, the model's ability to handle diverse tasks makes it a powerful tool for both technical and creative projects. Here are more guides from our previous articles and guides related to Qwen AI that you may find helpful. Performance Benchmarks Qwen 3.8 Max has demonstrated exceptional performance across various industry benchmarks, solidifying its position as a leader in AI innovation. Key achievements include: * Code Design: Ranked fourth in the LM Arena, competing closely with other top-tier models like Opus 4.8 and Fable, showcasing its strength in coding and design tasks. * Agentic Tasks: Excels in structured reasoning and tool usage, further establishing its reputation as a versatile and reliable AI model. * Python Execution: Delivers precise and efficient results, making it a valuable resource for developers tackling complex programming challenges. Real-World Applications The practical applications of Qwen 3.8 Max highlight its versatility and potential to bridge technical and creative domains. Some notable examples include: * Space Tracking: Developed a web application to track the International Space Station in real-time, demonstrating its ability to handle data-intensive tasks. * Creative Projects: Designed a comprehensive Pokémon encyclopedia complete with animations, showcasing its creative potential. * Interactive Tools: Built an interactive tourist map for Los Angeles, emphasizing its capacity to generate engaging and user-friendly outputs. These examples illustrate how Qwen 3.8 Max can cater to diverse industries, from scientific research to entertainment and tourism, making it a valuable tool for a wide range of users. Accessibility and Pricing Qwen 3.8 Max offers competitive API pricing, making it an attractive option for businesses seeking advanced AI capabilities without incurring excessive costs. Additionally, the upcoming Qwen 3.8 27B model is designed to run on consumer hardware, significantly broadening access to innovative AI technology for individual users and small businesses. This widespread access of AI ensures that even smaller organizations can use its powerful features. Challenges to Consider While Qwen 3.8 Max is a remarkable achievement, it is not without its challenges. Some of the key considerations include: * Licensing Uncertainty: The model's licensing terms have yet to be fully disclosed, raising questions about its commercial use and potential restrictions. * Enterprise Focus: Its large-scale design primarily targets enterprise applications, which may limit its appeal to general users. However, the smaller Qwen 3.8 27B model is expected to address this limitation by offering a more accessible solution for smaller-scale applications. These challenges highlight the importance of continued development and refinement to ensure the model meets the needs of a broader audience. The Road Ahead The release of Qwen 3.8 Max marks a significant milestone in the evolution of artificial intelligence. Its new features, including its open-weight design and advanced capabilities, set a new benchmark for what AI can achieve. Looking forward, the anticipated Qwen 3.8 27B model promises to further provide widespread access to access to advanced AI, making it more accessible to individual users and small businesses. As open-weight models like Qwen continue to evolve, they are poised to play a pivotal role in shaping the future of technology. By driving innovation across industries and redefining how we interact with AI, Qwen 3.8 Max and its successors are set to leave a lasting impact on the technological landscape. Media Credit: Prompt Engineering Disclosure: Some of our articles include affiliate links. If you buy something through one of these links, Geeky Gadgets may earn an affiliate commission. Learn about our Disclosure Policy.

Qwen 3.8-Max launches with 2.4 trillion parameters. Don't trust the benchmarks just yet

Alibaba launched its large AI model Qwen3.8-Max with many parameters. This model supports extensive context windows and multimodal inputs for complex tasks. OpenAI also hinted at its next model, Astra, for advanced problem-solving. Both companies are focusing on autonomous AI workers, not just chatbots. Independent evaluations are needed before making deployment decisions. Alibaba just launched a 2.4 trillion-parameter AI model. But that's not the biggest story. Alibaba has unveiled Qwen3.8-Max, its largest AI model yet with 2.4 trillion parameters, making it the second-largest publicly announced model globally after Moonshot AI's Kimi K3. The headline number grabbed attention and even pushed Alibaba's stock higher. But the parameter count isn't what matters most. The bigger questions are simple: How good is the model really? And should anyone trust the benchmarks yet? A giant model that doesn't use all of itself Despite having 2.4 trillion parameters, Qwen3.8-Max isn't activating the entire network every time it generates text. It's a Mixture-of-Experts (MoE) model, meaning only about 95 billion parameters are active per token. That keeps inference costs significantly lower than running a traditional dense model of comparable size while still delivering frontier-level capabilities. Alibaba says the model supports a one million token context window (roughly 750,000 words, allowing it to process entire books or large codebases in a single prompt), multimodal inputs (meaning it can understand text, images, documents and videos together), and can generate outputs of up to 131,072 tokens (around 100,000 words) in one response. The company has also promised to release open weights next week, allowing developers to download and run the model themselves instead of accessing it only through Alibaba's cloud, alongside a much smaller 27-billion-parameter (27B) version that is likely to be far more practical for enterprises to deploy on their own infrastructure. The benchmarks come with a big disclaimer Alibaba published an extensive benchmark comparison against models from OpenAI and Anthropic. The problem? Alibaba ran every single benchmark itself. That means it chose the prompts, sampling methods, retry policies and evaluation settings. At launch, there were no independent scores from platforms such as Artificial Analysis or community leaderboards. So every benchmark should be treated as a vendor claim rather than an independently verified result. Interestingly, Alibaba didn't try to paint a flawless picture. The company shows Qwen leading on benchmarks such as PaperBench and IFBench, suggesting improvements in long-horizon reasoning and instruction following. But it also openly shows the model trailing Anthropic's Claude on SWE-bench Pro, one of the industry's most important software engineering evaluations, as well as Humanity's Last Exam, a benchmark designed to measure advanced reasoning. The real pitch isn't benchmarks Alibaba appears to be positioning Qwen3.8-Max less as a chatbot and more as an autonomous AI worker. The company demonstrated the model spending 16 days building a command-line project on its own, producing hundreds of commits, pull requests and issues. In another demonstration, it reproduced an academic machine learning paper over several days, while a third experiment placed the model in a 24-hour data science competition where it reportedly outperformed most participating human teams. Unlike many AI demos, Alibaba has published the repositories, allowing developers to inspect what the model actually did rather than simply watching a polished video. Why Indian developers should care For engineering teams, GCCs and AI startups in India, pricing could matter far more than benchmark rankings. Alibaba has priced Qwen significantly below premium Western models, with additional discounts for cached inputs. That's particularly relevant for agentic workloads, where long-running AI systems repeatedly reuse the same context, making cache costs a major part of overall inference expenses. The caveat is reasoning. The model supports an enormous reasoning budget, but those reasoning tokens are billed as output tokens. Teams that leave deep reasoning enabled for routine workloads could end up paying substantially more than expected. Pricing is another area where Alibaba is trying to stand out. According to its official Model Studio rate card, Qwen3.8-Max costs $2 per million input tokens (the text you send to the model) and $6 per million output tokens (the text the model generates). Cached reads, where the model reuses previously processed context instead of reading it from scratch, cost just $0.25 per million tokens using implicit caching and $0.17 with explicit caching. More importantly, Alibaba applies the same pricing across the model's entire one-million-token context window, without charging extra for very long prompts. Most frontier AI models increase costs as prompts become longer, making this flat pricing structure an unusual and potentially significant advantage for developers building long-running AI agents. OpenAI quietly revealed its next move too Almost unnoticed, OpenAI also hinted at its next flagship model. Instead of a product launch, the company mentioned Astra in the third paragraph of a mathematics research blog, claiming the model had generated new results across several long-standing problems in mathematics and theoretical computer science. Like Alibaba, OpenAI's evidence is impressive but largely self-produced. While the mathematical proofs are publicly available for verification, researchers have questioned how many problems were attempted, what role humans played, and how the experiments were conducted. The bigger trend Looking past the headlines, both Alibaba and OpenAI are pushing the industry toward the same destination. Neither company is talking about faster chatbots anymore. Instead, they're trying to convince developers that their models can work independently for hours or even days on complex tasks. The challenge is that both companies are also asking the industry to trust evidence they've generated themselves. Independent evaluations for Qwen are expected soon, while Astra may take much longer to assess. Until then, the safest approach for enterprises remains the same: ignore the marketing tables, wait for third-party benchmarks, and test the models on your own workloads before making deployment decisions.

Alibaba Launches Qwen 3.8 Max with a 1 Million Token Context

Alibaba's Qwen 3.8 Max has been released, featuring 2.4 trillion parameters and a focus on advanced tasks such as multimodal workflows and 3D simulation generation. The open-weight Qwen 3.8 27B model complements it by allowing developers to customize deployments locally, offering greater control over implementation. According to World of AI, while these models bring notable strengths in areas like front-end development and long-horizon reasoning, they face challenges in competing with higher-ranking models on certain benchmarks. Discover how Qwen 3.8 Max performs in 3D asset creation, integrates visual and textual data for multimodal tasks and supports cost-efficient large-scale projects. Gain insight into its specialized strengths, as well as its limitations in handling simpler workflows where speed and precision are prioritized. This overview provides a detailed evaluation to help you determine whether it aligns with your specific requirements. Qwen 3.8 Max Qwen 3.8 Max is one of the most ambitious AI architectures developed to date. With its 2.4 trillion parameters, it is designed to tackle complex tasks, including coding, research, and multimodal workflows. Its smaller sibling, Qwen 3.8 27B, offers an open-weight deployment model, allowing users to run it locally for enhanced customization and control. Both models are optimized for long-horizon reasoning, making them well-suited for workflows requiring advanced problem-solving and integration of diverse data types. The open-weight nature of Qwen 3.8 27B is particularly noteworthy, as it provides developers and researchers with the flexibility to adapt the model to specific needs without relying on external infrastructure. This feature underscores Alibaba's commitment to fostering innovation by empowering users with greater autonomy. Performance: Strengths and Challenges Qwen 3.8 Max's architecture is undeniably innovative, but its performance metrics reveal a mixed picture. Ranking 12th in AI performance benchmarks, it lags behind leading competitors like Opus 4.7 and DeepSeek 4 Flash. However, its strengths lie in specialized areas, including: * Front-end development for creating interactive web applications * 3D asset generation, such as architectural models and simulations * Interactive workflows that integrate multimodal data Despite these strengths, the model struggles with simpler tasks. Its tendency to "overthink" can result in slower execution times, making it less practical for routine applications or straightforward coding tasks. Additionally, while its output quality is functional, it lacks the refinement and polish seen in newer models like Claude Opus 5 or DeepSeek 4 Flash, which excel in both speed and precision. Advance your skills in Qwen AI models by reading more of our other detailed content. Core Capabilities and Applications Qwen 3.8 Max's standout feature is its multimodal functionality, which allows it to process and analyze visual data alongside textual inputs. This capability is particularly valuable for projects requiring the integration of diverse data types. Key applications include: * 3D simulation generation, such as creating architectural designs, engineering prototypes, or even solar system models * End-to-end workflows that demand seamless handling of text, images and other data formats These capabilities make Qwen 3.8 Max a versatile tool for developers and researchers working on data-intensive projects. For instance, it has been successfully used to create browser-based operating system clones and interactive simulations, demonstrating its potential to drive innovation in digital environments. Affordability and Cost-Effectiveness One of Qwen 3.8 Max's most appealing features is its competitive pricing structure, which makes it accessible for a wide range of users. The model offers: * Input tokens priced at $2 per 1 million * Output tokens priced at $6 per 1 million With a 1 million token context window, Qwen 3.8 Max can process extensive datasets without incurring prohibitive costs. This affordability is particularly advantageous for large-scale projects or users operating within budget constraints. By balancing cost with functionality, Qwen 3.8 Max positions itself as a practical choice for organizations seeking high-capacity AI solutions without overspending. Competitive Landscape In the competitive AI market, Qwen 3.8 Max offers a unique blend of affordability, flexibility, and specialized capabilities. However, it faces stiff competition from high-performing models like Claude Opus 5, which delivers superior output quality and DeepSeek 4 Flash, known for its speed and cost-efficiency. While Qwen 3.8 Max may not lead in overall performance, its open-weight deployment options and cost-effectiveness provide distinct advantages for users prioritizing customization and budget management. For developers and researchers seeking a balance between performance and affordability, Qwen 3.8 Max offers a compelling alternative. Its ability to integrate into diverse workflows and handle complex tasks makes it a valuable tool, even if it falls short of the top-tier models in certain areas. Future Potential The open-weight Qwen 3.8 27B model holds significant promise for local deployment, offering users greater control over their projects. This feature is particularly appealing for organizations prioritizing data privacy and customization. Future updates to the Qwen 3.8 Max architecture could address its current limitations, such as efficiency in simpler tasks and output refinement, potentially enhancing its appeal in the competitive AI landscape. As the technology evolves, Qwen 3.8 Max has the potential to carve out a more prominent role in the AI market. Its focus on multimodal functionality, 3D simulation generation, and cost-effective solutions positions it as a versatile option for specialized use cases. With continued development, it could become a more competitive and well-rounded choice for a broader range of applications. Media Credit: WorldofAI Disclosure: Some of our articles include affiliate links. If you buy something through one of these links, Geeky Gadgets may earn an affiliate commission. Learn about our Disclosure Policy.

Alibaba AI model: Alibaba unveils its most capable AI model to date, not far behind Moonshot's in size

Chinese tech companies - a huge force in open-weight AI models globally - are locked in a fierce and fast-moving battle to build more powerful systems without making them prohibitively ⁠expensive ⁠to run. China's Alibaba on Monday unveiled what it said is its largest and most capable artificial-intelligence model, the Qwen3.8-Max, which is not far behind in size when compared with an offering from domestic rival Moonshot AI launched last month. Chinese tech companies - a huge force in open-weight AI models globally - are locked in a fierce and fast-moving battle to build more powerful systems without making them prohibitively ⁠expensive ⁠to run. Qwen3.8-Max has 2.4 trillion parameters, the numerical settings a model learns from data and uses to recognise patterns, generate answers, and carry out tasks. Moonshot's Kimi K3 has 2.8 trillion parameters. A higher figure does not automatically make a model better, but it has become a closely watched measure of the scale of the computing and data behind advanced AI systems. Chinese tech companies are keen to publish parameter count to help their models gain ⁠traction among the developer community. Their models tend to be open-weight, meaning the underlying learned settings that allow developers to run or adapt the system are available for ⁠download. By contrast, OpenAI, Anthropic and Google do not publish parameter count for their closed-source models. Qwen3.8-Max was unveiled on crowdsourced, model-comparison platform Arena.AI, where it immediately became the highest-ranking Chinese model in terms of text models, though it still lags Claude Fable 5 and three Opus variants which are all from Anthropic. But on Arena.AI's leaderboard for AI models that analyse images and other visual material, Qwen3.8-Max ranked second globally, only behind a Claude Fable 5 variant. Both Qwen3.8-Max and Kimi K3 can handle text, images and video, and process up to 1 million tokens at a time. Tokens are chunks of data, often parts of words or short words, and a big figure means ⁠the model can take in large amounts of material in one go, such as long legal files, a large software codebase or hundreds of pages of documents. Alibaba said its model uses a "mixture-of-experts" design, which divides work among specialised parts of the system instead of switching on the entire model for every request. Only 95 billion parameters are used at a time, reducing costs and response delays. The tech giant said the model completed a software-engineering project in 16 days. The Qwen3.8-Max is due to be released next week through Alibaba Cloud's Model Studio platform.

Alibaba Launches Qwen3.8 Max to Challenge Anthropic AI

Alibaba has launched Qwen3.8 Max, its biggest AI model yet, with 2.4 trillion parameters. The Chinese tech giant announced the model on August 3 as China's AI race grows stronger. Alibaba says Qwen3.8 Max can match or beat Anthropic's Fable 5 on several tests. The model also ranks above Moonshot AI's Kimi K3 on several benchmarks. Qwen3.8 Max supports text, images, videos, and other large files through its multimodal AI abilities. says the model can handle one million tokens in a single context window. The system also performed strongly in coding and completed a software project independently during a 16-day internal test. Alibaba uses a design that activates only part of the model for each task, helping reduce computing costs. The company also plans to release Qwen3.8 Max weights publicly during the week starting August 10. Open access could help developers download, change, and build new applications with the model. Ling Vey-Sern, managing director at Union Bancaire Privee, said, "The gap is probably much closer, and narrowing fast." He described as another sign of China's fast AI progress. The launch adds fresh pressure to Anthropic, OpenAI, and other major AI companies. Alibaba also faces growing competition from Chinese firms such as Moonshot AI, DeepSeek, Z.ai, and ByteDance. The latest release shows how quickly the China AI race continues moving forward. Official Alibaba Qwen X post: The current coverage confirms Alibaba announced through its official Qwen social channels, with open weights planned for next week.

Why is Alibaba stock surging today? By Investing.com

Investing.com -- Alibaba's Hong Kong stock surged 6.4% to HK$124.5 after the company released a new, advanced artificial intelligence model. The technology giant unveiled Qwen 3.8-MAX, the latest version of its flagship artificial intelligence model, which it said delivered improvements in reasoning, coding, agent capabilities and multimodal understanding, while offering lower inference costs than previous versions. Adding to the positive sentiment, Bloomberg reported Chinese AI startup Moonshot AI has a computing power agreement with Alibaba to access a cluster of approximately 20,000 Nvidia chips -- a substantial portion of the total compute capacity underpinning Moonshot's Kimi AI models, including the recently unveiled Kimi K3 system. The arrangement ties Alibaba more tightly to one of China's most advanced AI model developers, with Moonshot's Kimi models relying on Alibaba Cloud as a core part of their computing infrastructure -- reinforcing Alibaba's role not only as an e-commerce and digital services provider, but also as a key infrastructure partner for high-compute AI workloads. Alibaba is also one of Moonshot's largest investors. Broader Asian technology stocks also advanced, recovering from deep losses in July. This article was generated with the support of AI and reviewed by an editor. For more information see our T&C.

Alibaba unveils Qwen 3.8-MAX AI model; shares jump By Investing.com

Investing.com-- Alibaba Group (HK:9988) shares rose on Monday after the technology giant unveiled Qwen 3.8-MAX, the latest version of its flagship artificial intelligence model, as competition intensifies among Chinese firms racing to develop more powerful generative AI systems. Hong Kong-listed Alibaba shares advanced 6% to HK$124.00 by 02:58 GMT. Alibaba said that Qwen 3.8-MAX delivers improvements in reasoning, coding, agent capabilities and multimodal understanding, while offering lower inference costs than previous versions. Get breaking news on key AI-related developments with InvestingPro -- at 60% off now The new model, built on the Qwen 3.5 architecture, features 2.4 trillion parameters with 95 billion active parameters and will become the first Max-class Qwen model whose weights will be released as open source next week, Alibaba said. Alibaba said benchmark tests showed Qwen3.8-Max was broadly competitive with leading U.S. AI models from OpenAI and Anthropic, outperforming them on several coding, multimodal and engineering benchmarks while trailing on some general-purpose reasoning tests. The launch comes as Alibaba continues to ramp up investment in AI infrastructure and cloud computing to compete with domestic rivals including DeepSeek, Baidu and Tencent, while also challenging leading U.S. models. Earlier in May, Alibaba said it would exceed its planned AI investment of up to 380 billion yuan ($55.96 billion) over the next three years. The rollout also follows a series of rapid model upgrades by Chinese AI companies as competition accelerates after DeepSeek's low-cost reasoning models reshaped the industry's competitive landscape earlier this year.

Alibaba unveils its largest AI model yet, DeepSeek's latest model is ultra-low cost

BEIJING, Aug 3 (Reuters) - China's Alibaba on Monday unveiled its largest and most capable AI model to date, sending its shares surging, while a research firm said DeepSeek's latest product offers cut-throat pricing that is more than 100 times cheaper than Anthropic's Claude Fable 5. The two developments highlight the rapid pace of advancement in artificial intelligence by Chinese tech firms, which are locked in a fierce and fast-moving battle to build more powerful systems without making them prohibitively expensive to run. Both models -- Alibaba's Qwen3.8-Max and DeepSeek's V4-Flash -- underline Chinese commitment to open-weight models as the firms seek to gain traction among developers globally. "Chinese AI companies have found an important market. Many business workflows do not need the industry's very best model," said Lian Jye Su, chief analyst at research firm Omdia. "They need models that are good enough, affordable, transparent and accessible, and open-weight models help meet that demand." With an open-weight model, the underlying learned settings that allow developers to run or adapt the system are available for download. By contrast, OpenAI, Anthropic and Google have closed-source models. TRILLIONS OF PARAMETERS Alibaba's new Qwen3.8-Max immediately shot up leaderboards assessing the capabilities of AI models after being unveiled on Monday, helping its shares jump 7% in Hong Kong trade. The model has 2.4 trillion parameters, the numerical settings a model learns from data and uses to recognise patterns, generate answers, and carry out tasks. That puts it not too far behind domestic rival Moonshot AI's Kimi K3, which was launched last month and has 2.8 trillion parameters. A higher parameter figure does not automatically make a model better, but it has become a closely watched measure of the scale of the computing and data behind advanced AI systems. Qwen3.8-Max was unveiled on crowdsourced, model-comparison platform Arena.AI. It soon became the highest-ranking Chinese model in terms of text models, though it still lags Claude Fable 5 and three Opus variants which are all from Anthropic. On Arena.AI's leaderboard for AI models that analyse images and other visual material, Qwen3.8-Max ranked second globally, only behind a Claude Fable 5 variant. Both Qwen3.8-Max and Kimi K3 can handle text, images and video, and process up to 1 million tokens at a time. Tokens are chunks of data, often parts of words or short words, and a big figure means the model can take in large amounts of material in one go, such as long legal files, a large software codebase or hundreds of pages of documents. The tech giant said the model, due to be released next week, completed a software-engineering project in 16 days. It uses a "mixture-of-experts" design, which divides work among specialised parts of the system instead of switching on the entire model for every request. Only 95 billion parameters are used at a time, reducing costs and response delays. DEEPSEEK IS ULTRA CHEAP DeepSeek's V4-Flash model, released on Friday, is by far the least expensive to run on benchmark tests among well-known models globally, according to research firm Artificial Analysis. The startup, which sources have said is preparing for a potential IPO, saw its R1 and V3 models become a global sensation in early 2025, triggering a selloff in global tech stocks and raising questions about the large amounts U.S. companies were spending on AI. V4-Flash charges $0.14 per million input tokens and $0.28 per million output tokens, according to San Francisco-based Artificial Analysis. Artificial Analysis estimated V4-Flash's average cost at 3 cents per test, compared with 86 cents for Kimi K3, $1.86 for OpenAI's GPT-5.6 Sol and $3.15 for Claude Fable 5. The comparison provides a more realistic measure of value than pricing alone because it accounts for the amount of data a model must process and generate to complete a task. A model with low headline price can still prove expensive if it requires significantly more steps to produce an answer. (Reporting by Eduardo Baptista; Editing by Edwina Gibbs)

Alibaba Releases New AI Model as Competition Intensifies

China's Alibaba Group released a new artificial-intelligence model as technology giants continue to race to build ever more capable systems. The Hangzhou-based company said the model, called Qwen 3.8-Max, is the "most powerful model in the Qwen series to date." The model boasts 2.4 trillion parameters and will be fully open-source next week. AI system parameters work like brain cells: The more a model has, the more knowledge it can store, making the count a shorthand for a model's capabilities. The model, available on Alibaba's developer platforms, delivers a comprehensive upgrade across coding, real-world work, research and long-horizon tasks. Operating as a multimodal foundation, it also supports visual intelligence, the Chinese company said. Qwen 3.8-Max ranks fifth on the Text Arena leaderboard for text-to-text tasks across math, coding and creative writing, behind several models by Anthropic. It is No. 2 on Vision Arena, which rates a model's ability to reason over visual inputs, according to rankings compiled by Alibaba. China's rapid advances in AI have been spurred in part by Beijing's push for self-sufficiency as it competes for technological supremacy against the U.S. China's AI startup Moonshot AI recently released its Kimi K3 model, which has 2.8 trillion parameters. Alibaba has been betting on AI and cloud as its growth driver for the next phase. The company expects AI-related product revenue to become the primary engine of revenue growth for the cloud segment, Chief Executive Eddie Wu said earlier this year.

What is Alibaba's Qwen 3.8-Max? How it is different from Claude Fable 5 and Kimi K3

Alibaba says Qwen 3.8-Max rivals Kimi K3 while positioning it just behind Anthropic's Claude Fable 5 in global AI benchmarks. Alibaba has introduced its latest Qwen 3.8-Max, the most advanced AI model till date. This new model directly rivals AI systems such as Anthropic's Claude Fable 5 and Moonshot AI's Kimi K3. The company stated that the new model is made for complex reasoning, autonomous coding and enterprise grade AI workloads while also becoming the first flagship model in the Qwen lineup that will eventually be released with open weights. Built on a 2.4 trillion-parameter architecture Qwen 3.8-Max is based on a Mixture-of-Experts (MoE) architecture featuring 2.4 trillion total parameters, with around 95 billion active parameters used for each token. This approach is intended to reduce computing costs while maintaining high performance. The model is natively multimodal, allowing it to process text, images and videos, and supports a 1 million-token context window, enabling it to analyse lengthy documents and conversations in a single session. Also read: Apple iPhone Ultra might debut next month, may compete with Samsung Galaxy Z Fold 8: Check expected specs and price Focus on coding and enterprise AI The company said that Qwen 3.8-Max is optimised for long running AI tasks and autonomous software development. As per the company, the model can independently write code, identify bugs, run tests and fix errors over extended periods without continuous human input. Apart from programming, the model is designed for enterprise applications including legal document analysis, data processing, research workflows and advanced document reasoning. How it compares with rivals Alibaba has positioned Qwen 3.8-Max as one of the strongest AI models currently available, though it acknowledges that Anthropic's Claude Fable 5 still leads global benchmark rankings. If we compare it to Kimi K3, Qwen 3.8-Max offers similar large-context capabilities but differentiates itself through its planned open-weight release, allowing developers to deploy and customise the model locally. Meanwhile, Claude Fable 5 remains a closed source commercial offering focused on advanced reasoning and coding performance.