Alibaba’s Qwen Team Launches Qwen3.7-Plus, Adding Vision, Deep Reasoning, Tool Invocation, and Autonomous Iteration on the Bailian Platform

重點摘要
Alibaba’s Qwen team has released Qwen3.7-Plus. The model is now available through Alibaba Cloud’s Bailian platform. Bailian is the console international users access as Model Studio. It offers API services to external developers. The release follows Alibaba’s May unveiling of the Qwen3.7 generation. Qwen3.7-Plus Qwen3.7-Plus is a multimodal large language model. The model understands images and video, alongside written prompts. Its sibling, Qwen3.7-Max, is text-only. This is visual understanding, not generation. The model reads images and video; it does not create them. Alibaba’s image and video generation work sits in separate model families. Alibaba team describes the release as a step in multimodal hybrid agent technology. An agent is a model that plans and acts across steps. Building o
Alibaba’s Qwen team has released Qwen3.7-Plus. The model is now available through Alibaba Cloud’s Bailian platform. Bailian is the console international users access as Model Studio. It offers API services to external developers. The release follows Alibaba’s May unveiling of the Qwen3.7 generation. Qwen3.7-Plus Qwen3.7-Plus is a multimodal large language model. The model understands images and video, alongside written prompts. Its sibling, Qwen3.7-Max, is text-only. This is visual understanding, not generation. The model reads images and video; it does not create them. Alibaba’s image and video generation work sits in separate model families. Alibaba team describes the release as a step in multimodal hybrid agent technology. An agent is a model that plans and acts across steps. Building on image and video understanding, Qwen3.7-Plus adds five abilities. These are deep reasoning, self-programming, tool invocation, verification and testing, and autonomous iteration. Self-programming means the model writes and revises its own code. Tool invocation means it calls external functions or APIs. Verification and testing means it runs outputs and checks results. Autonomous iteration means it loops until the task is done. Together, they describe a model built to act, not just answer. The Vision Case Qwen3.7-Plus is the multimodal half of the 3.7 family. Its preview already posted measurable vision results. In Vision Arena, Qwen3.7-Plus-Preview ranked #16 overall. That placed Alibaba as the #5 lab in vision. The model rank and the lab rank are separate figures. Vision Arena is a neutral leaderboard run by LM Arena. Users vote on image-understanding answers in blind matchups. The #16 result sits behind the top US labs, but inside the field. For image-heavy work, this is the signal that matters. Think OCR at scale, chart reading, or video-frame analysis. The text-only Max sibling anchors the generation’s reasoning. Max scored 56.6 on the Artificial Analysis Intelligence Index. That was the highest placement for a Chinese model at release. https://qwen.ai/blog?id=qwen3.7-plus The Agentic Loop The clear shift in Qwen3.7 is its agentic focus. Alibaba team is positioning the models for long-running tasks. Bailian, the host platform, adds two relevant pieces. The first is an Agentic RL (reinforcement learning) mechanism. The platform uses real-world execution feedback to refine model accuracy over time. The second is a set of built-in safety guardrails. These keep autonomous tools inside preset operational limits. That detail matters when an agent runs commands or edits files. Marktechpost’s Visual Explainer #mtp-qwen37plus-slider.mtp-root{ --mtp-bg:#ffffff; --mtp-canvas:#f4f5f0; --mtp-ink:#15171a; --mtp-sub:#565b61; --mtp-line:#e4e6df; --mtp-green:#76B900; --mtp-green-ink:#4d7a00; --mtp-soft:#f0f3e8; --mtp-serif:'Fraunces',Georgia,'Times New Roman',serif; --mtp-sans:'Inter',-apple-system,BlinkMacSystemFont,'Segoe UI',sans-serif; --mtp-mono:'JetBrains Mono','SFMono-Regular',Menlo,Consolas,monospace; } @import url('https://fonts.googleapis.com/css2?family=Fraunces:opsz,[email protected],400;9..144,500;9..144,600&family=Inter:wght@400;500;600&family=JetBrains+Mono:wght@500;600&display=swap'); #mtp-qwen37plus-slider, #mtp-qwen37plus-slider *{box-sizing:border-box!important;margin:0;padding:0} #mtp-qwen37plus-slider hr,#mtp-qwen37plus-slider p:empty,#mtp-qwen37plus-slider del,#mtp-qwen37plus-slider s{display:none!important} #mtp-qwen37plus-slider.mtp-root{ background:var(--mtp-canvas)!important; color:var(--mtp-ink)!important; font-family:var(--mtp-sans); border:1px solid var(--mtp-line)!important; border-radius:18px; padding:22px; max-width:860px; margin:24px auto; line-height:1.55; -webkit-font-smoothing:antialiased; } #mtp-qwen37plus-slider .mtp-card{ background:var(--mtp-bg)!important; border:1px solid var(--mtp-line)!important; border-radius:14px; overflow:hidden; box-shadow:0 1px 0 rgba(21,23,26,.02),0 18px 40px -28px rgba(21,23,26,.25); } #mtp-qwen37plus-slider .mtp-bar{ display:flex;align-items:center;justify-content:space-between; padding:16px 22px;border-bottom:1px solid var(--mtp-line)!important; background:var(--mtp-bg)!important; } #mtp-qwen37plus-slider .mtp-brand{ font-family:var(--mtp-mono);font-size:12px;letter-spacing:.14em; text-transform:uppercase;color:var(--mtp-green-ink)!important;font-weight:600; } #mtp-qwen37plus-slider .mtp-count{ font-family:var(--mtp-mono);font-size:12px;color:var(--mtp-sub)!important;font-weight:600; } #mtp-qwen37plus-slider .mtp-count b{color:var(--mtp-ink)!important} #mtp-qwen37plus-slider .mtp-stage{position:relative;padding:34px 30px 30px;background:var(--mtp-bg)!important;min-height:300px} #mtp-qwen37plus-slider .mtp-slide{display:none} #mtp-qwen37plus-slider .mtp-slide.is-on{display:block;animation:mtpIn .45s cubic-bezier(.2,.7,.2,1)} @keyframes mtpIn{from{opacity:0;transform:translateY(10px)}to{opacity:1;transform:none}} #mtp-qwen37plus-slider .mtp-kick{ font-family:var(--mtp-mono);font-size:12px;letter-spacing:.12em;text-transform:uppercase; color:var(--mtp-green-ink)!important;font-weight:600;display:inline-block;margin-bottom:14px; } #mtp-qwen37plus-slider .mtp-kick::before{content:"";display:inline-block;width:22px;height:2px;background:var(--mtp-green)!important;vertical-align:middle;margin-right:9px;border-radius:2px} #mtp-qwen37plus-slider h2.mtp-h{ font-family:var(--mtp-serif);font-weight:500;font-size:30px;line-height:1.15;color:var(--mtp-ink)!important;margin-bottom:16px;letter-spacing:-.01em; } #mtp-qwen37plus-slider .mtp-lead{font-size:16px;color:var(--mtp-sub)!important;max-width:60ch} /* cover */ #mtp-qwen37plus-slider .mtp-eyebrow{font-family:var(--mtp-mono);font-size:12px;letter-spacing:.16em;text-transform:uppercase;color:var(--mtp-green-ink)!important;font-weight:600;margin-bottom:18px} #mtp-qwen37plus-slider h1.mtp-title{font-family:var(--mtp-serif);font-weight:600;font-size:48px;line-height:1.05;color:var(--mtp-ink)!important;letter-spacing:-.02em;margin-bottom:10px} #mtp-qwen37plus-slider .mtp-title small{display:block;font-family:var(--mtp-serif);font-style:italic;font-weight:400;font-size:22px;color:var(--mtp-sub)!important;margin-top:8px;letter-spacing:0} #mtp-qwen37plus-slider .mtp-meta{margin-top:22px;padding-top:16px;border-top:1px solid var(--mtp-line)!important;font-size:14px;color:var(--mtp-sub)!important} #mtp-qwen37plus-slider .mtp-meta b{color:var(--mtp-ink)!important;font-weight:600} #mtp-qwen37plus-slider .mtp-hint{margin-top:14px;font-family:var(--mtp-mono);font-size:12px;color:var(--mtp-green-ink)!important} /* lists */ #mtp-qwen37plus-slider ul.mtp-list{list-style:none;margin-top:6px} #mtp-qwen37plus-slider ul.mtp-list li{position:relative;padding:11px 0 11px 26px;border-bottom:1px solid var(--mtp-line)!important;font-size:15.5px;color:var(--mtp-ink)!important} #mtp-qwen37plus-slider ul.mtp-list li:last-child{border-bottom:0!important} #mtp-qwen37plus-slider ul.mtp-list li::before{content:"";position:absolute;left:2px;top:18px;width:8px;height:8px;border-radius:50%;background:var(--mtp-green)!important} #mtp-qwen37plus-slider ul.mtp-list li b{font-weight:600} #mtp-qwen37plus-slider ul.mtp-list li span{color:var(--mtp-sub)!important} #mtp-qwen37plus-slider .mtp-note{margin-top:16px;padding:12px 14px;background:var(--mtp-soft)!important;border-left:3px solid var(--mtp-green)!important;border-radius:6px;font-size:14px;color:var(--mtp-ink)!important} /* two columns */ #mtp-qwen37plus-slider .mtp-cols{display:grid;grid-template-columns:1fr 1fr;gap:16px;margin-top:6px} #mtp-qwen37plus-slider .mtp-col{border:1px solid var(--mtp-line)!important;border-radius:10px;padding:16px;background:var(--mtp-bg)!important} #mtp-qwen37plus-slider .mtp-col h3{font-family:var(--mtp-mono);font-size:11px;letter-spacing:.1em;text-transform:uppercase;font-weight:600;margin-bottom:10px;color:var(--mtp-ink)!important} #mtp-qwen37plus-slider .mtp-col.ok h3{color:var(--mtp-green-ink)!important} #mtp-qwen37plus-slider
Related
相關文章
Liquid AI Introduces LFM2.5-Embedding-350M and LFM2.5-ColBERT-350M: Dense Bi-Encoder and Late-Interaction Models for Fast Multilingual Search Across 11 Languages
This week, Liquid AI released two new retrieval models. They are LFM2.5-ColBERT-350M and LFM2.5-Embedding-350M. Both hold 350M parameters. Both are the first bidirectional members of the LFM family. They build on LFM2.5-350M-Base, released in March. The pair targets fast multilingual and cross-lingual search across 11 languages. Their footprint is small enough to run almost anywhere. Both are available now on Hugging Face under the LFM Open License v1.0. LFM2.5 Retrievers The two models share one backbone but represent text differently. LFM2.5-Embedding-350M is a dense bi-encoder. It turns each document into a single vector. Pick it when you want the fastest search and the smallest, cheapest index. LFM2.5-ColBERT-350M is a late-interaction model. It converts each token into a vector rather
Perplexity Launches Brain, a Self-Improving Memory System That Builds a Context Graph of an Agent’s Work and Learns Overnight
Most AI memory remembers the user. It stores your preferences, your tastes, and your role. Perplexity is taking a different path. Today, Perplexity launched Brain, a self-improving memory system for its agent product, Computer. Brain does not focus on remembering you. It remembers what the agent did. That reframes what memory in AI is for. What is Perplexity‘s Brain Brain is a self-improving memory system. It builds a context graph of the work Computer performs. At set intervals, such as overnight, Brain reviews that graph. It then teaches itself how to do the work better. The idea is straightforward. The more work you do, the more efficient Brain makes your Computer. Brain is rolling out today to Perplexity Max and Enterprise Max subscribers in Research Preview. Two Axes of AI Memory Perp

智譜新高,MiniMax承壓,“大模型雙雄”命運殊途
這篇消息聚焦「智譜新高,MiniMax承壓,“大模型雙雄”命運殊途」。原始導語提到:大模型在被市場重新定價 從 AI 情報角度來看,這類內容值得關注其背後的技術進展、產品落地、產業競爭與後續市場影響。

華為昇騰 0 Day 支持智譜 GLM-5.2 模型,提供全面推理優化
華為昇騰 AI 宣佈在智譜開源 GLM-5.2 大模型當天即完成深度推理優化。通過 MOE 大融合算子、通信計算融合、高併發調度等七項關鍵技術,顯著提升編程和長程任務的處理效率,現已支持 A3 系列產品部署。#AI 大模型# #國產算力#
企業AI轉型再添利器:青雲科技算力雲接入 MiniMax-M3 模型
企業AI落地面臨高效低成本難題。青雲科技旗下基石智算平臺接入國產開源大模型MiniMax-M3,提供新算力支持。MiniMax-M3以卓越上下文處理能力等三大核心技術見長,依託自研架構,助企業便捷部署AI業務。
阿里開源統一科學大模型 LOGOS,僅用五十六分之一參數超越微軟
阿里 ATH-Token Foundry 聯閤中國人民大學高瓴人工智能學院開源科學基礎模型 LOGOS。該模型採用統一科學語法與純序列建模範式,在六大科學任務上匹配或超越傳統專用方法。其中 LOGOS-1B 僅 1B 參數,即展現出極高效率,性能超越參數量達 8×7B 的微軟模型。