Alibaba's Qianwen Office adds GLM-5.3 and DeepSeek V4 Pro

Share
Alibaba's Qianwen Office adds GLM-5.3 and DeepSeek V4 Pro

Alibaba just turned its flagship office agent into a showroom for its rivals' models — a move no Western office suite has made yet.

Alibaba's Qianwen Office — the enterprise agent suite that opened public beta earlier this month — is the first product to offer Zhipu's GLM-5.3 and DeepSeek's V4 Pro directly, selectable today from the "frontier models" tier on the app's homepage. The two additions sit alongside Alibaba's own Qwen3.8, which shipped as the launch flagship when the product debuted in early August.

The launch is a quietly aggressive positioning play. Qianwen Office was assembled by merging three of Alibaba's agent products — QoderWork, MuleRun, and Wukong — and pitches itself as the industry's first suite to run desktop agents, cloud agents, and enterprise-collaboration agents from one product. Adding its two biggest domestic rivals' models a week into public beta turns that pitch into a genuine model-agnostic layer, and a live test of whether Chinese enterprises will pick best-of-breed brains over vendor lock-in.

The two additions complement each other. DeepSeek V4 Pro — whose open weights landed on Hugging Face just days ago — brings a 1M-token context window with up to 384K-token outputs, tuned for long-horizon agent chains and automation-heavy workflows. GLM-5.3, Zhipu's open-weights flagship, delivers a 50% performance lift over GLM-5.2 through post-training scaling alone, with standout results on long engineering tasks and white-box code review. Personal plans run from free to 198 yuan a month (78–158 yuan at launch discounts); enterprise seats go for 198 yuan per user per month.

The signal here is less about the models than the posture. Alibaba operates one of China's biggest model families, yet its flagship office product ships rivals' frontier models as first-class citizens — a walled-garden strategy would have kept them out. It's the same pattern we saw when Honor's YOYO Claw agent picked up GLM-5.3 last week, and it's the Chinese ecosystem's answer to the US model wars: platforms compete on workflow and distribution, not exclusivity, and open weights make swapping brains cheap. For enterprises the takeaway is choice — if Alibaba's own office agent runs DeepSeek and Zhipu models out of the box, the era of "pick one lab and live with it" is over.

What to watch: whether Qianwen Office grows into a full model marketplace with third-party pricing — and whether Tencent or ByteDance's office suites answer in kind.

If Alibaba's office suite ships rival models out of the box, how long before a Western vendor does the same? Tell us in the comments.

Read more

Byteification turns Qwen3 and Llama 3 into byte-level models

Byteification turns Qwen3 and Llama 3 into byte-level models

The tokenizer is the last piece of the language-model pipeline that was never learned end to end — and a Nature paper out this week shows it can be swapped out for less than one percent of a pretraining budget. Researchers have retrofitted four open subword language models into true byte-level models using less than 1% of a typical pretraining budget, and the result — published 7 October in Nature — closes the performance gap that has kept byte-level models a niche curiosity for nearly a decad

OpenAI's dots go mobile and can start your Codex work

OpenAI's dots go mobile and can start your Codex work

One product update and two benchmark papers this hour, all circling the same question: how much of the work can you actually hand over, and what happens to the part you didn't? OpenAI's dots — the always-on agents that get their own cloud computer — can now be created and configured entirely from the ChatGPT app on iOS and Android, and a dot can start Codex work or pick up an existing Codex thread on your behalf. The October 9 release, billed as the first of a fresh run of updates for dots, mo

Coding agents as a hospital: Cockroach's five-month results

Coding agents as a hospital: Cockroach's five-month results

The agentic-coding debate keeps being framed as speed versus trust. Cockroach Labs just published the most detailed trust-first ledger yet — five months of numbers, including the parts that didn't work. Cockroach Labs ran its codebase like a teaching hospital for five months and says the pipeline merged well over a million lines of code with seven total reverts. In the setup the company calls MOLT Sinai, GitHub issues are patients, merging is discharge, and the humans in charge are Chiefs of M

Today in AI — October 10, 2026

Today in AI — October 10, 2026

A day where the boring machinery took center stage: proof checkers, order books, warehouses and the humble text message — plus fresh arguments about who gets to declare any of it safe. Models & Research * Mathematicians get a reliability primer for the Lean Theorem Prover. A guest post by Thomas Hales on Terence Tao's blog walks through what Lean actually guarantees — and devotes itself to the "Summer of Soundness Bugs," the string of kernel bugs found this summer that let false proofs thr