Zhipu AI Latest News
Zhipu AI builds the GLM model family and enterprise AI services in China.
-
-
Sarvam 发布 Token Factory 推理服务:托管 GLM 5.3、DeepSeek V4.1 Flash 等多个开源权重模型
-
Revolut migrates about 2,000 engineers to self-hosted GLM for coding
-
Epoch AI Experiment Shows GLM 5.3 Can Autonomously Attack Firefox via Patch Gap
-
SGLang discloses and fixes cross-request KVCache race condition in PD-disaggregated serving
-
DeepSeek V4.1 Flash 1017 released, scores 45 on Artificial Analysis Intelligence Index to tie for second among open-source models
-
PromptArmor study finds open-weight models 88% more susceptible to indirect prompt injection
-
Goodfire trains and deploys production-grade cybersecurity monitors for Kimi K3 and GLM 5.3
-
Anthropic evaluates GLM-5.3: autonomously builds end-to-end network exploits, defenses easily bypassed
-
Fireworks AI launches FireRouter with Opus, cutting coding task costs by 57%
-
Artificial Analysis evaluates Step 5 Preview: Intelligence Index score of 44, cost about 1/2.8 of peer models
-
Arena updates Image-to-WebDev leaderboard: GPT-6 Astra tops with 1733
-
Together AI expands fine-tuning service with new models, experiment tracking and finer training control
-
Unsloth speeds up GLM-5.3-Flash local inference by up to 3.3x and releases GGUF
-
Baseten analyzes best open-source models for post-training and recommends them by cost tier
-
Zhipu open-sources GLM-5.3 model weights, focused on agentic coding and cyber defense
-
Unsloth releases guide for running quantized GLM-5.3 GGUF locally
-
GLM-5.3 releases open weights, strongest in agentic coding and cyber defense
-
Together AI benchmarks GLM-5.3 and GLM-5.3 Flash on DeepSWE for cost, coding and cascade routing
-
Unsloth releases GLM-5.3-Flash GGUF quantized local running solution
-
Tang Jie announces GLM-5.3 Flash AA tops OpenRouter
-
Together AI benchmarks GLM-5.3 against GPT-5.6 Sol: cost, coding and routing on DeepSWE
-
Baidu Qianfan lists GLM-5.3
-
GLM-5.3 launches: AA Intelligence Index score of 60 ties for first among open-source models at lower cost
-
Local small models can rival cloud frontier models: the alternative flight path of Qwen3.8-27B
-
GLM-5.3 pre-release safety evaluation: significant gains in cyber defense capabilities
-
ZCode upgrade adds Goal, Subagents, Remote Control and idle-time tasks
-
PromptArmor: vendors often follow up with international models such as Kimi, DeepSeek, and GLM after integrating inference providers like Fireworks
-
After wasting 2 billion tokens, I open-sourced Leader.skill to help agents define goals
-
After wasting 2 billion tokens, I open-sourced Leader.skill to help Agents define goals
-
Fireworks releases Fireworks Nexus: a routable open-source model intelligence layer for engineering teams
-
Kimi K3 Trails US Frontier Models Sharply in Cybersecurity Exploit Test, Knowledge Distillation May Be the Cause
-
UK AISI and CAISI jointly evaluate Kimi K3 cybersecurity capabilities, significantly below frontier models but above GLM-5.2
-
Build products without code: a step-by-step Vibe Coding tutorial from scratch
-
AISI's First Public Evaluation: Leading Open-Weight Models Lag Frontier by 4 to 7 Months in Cybersecurity Capabilities
-
Synthetic Sciences releases OpenScience: an open-source, model-agnostic AI workbench for machine learning, biology, physics and chemistry research
-
Zhipu launches ZCode, the official development environment for GLM-5.2
-
Artifacts 22: Zyphra, Cohere and Poolside are broadening the ecosystem
-
Weave launches intelligent model routing tool that plugs directly into Claude Code, Codex and Cursor
-
OpenRouter MCP server released
-
huggingface_hub achieves weekly releases: AI, open-source tools and human review in a closed loop
-
Modal releases Auto Endpoints: deploy a production-grade, self-controlled LLM inference service with one command
-
GLM-5.2: Built for Long-Horizon Tasks
-
GLM-5.2 Open-Sourced, Tops Global Benchmark and Matches Opus 4.8
-
GLM-5.2 launches and goes open-source, focused on coding and long-horizon tasks
-
Zhipu Releases Flagship GLM-5.2 with Stable 1M-Token Long-Horizon Task Capability
-
GLM-5.2 fully open: Zhipu's strongest open-source model supports 1M context
-
Zhipu GLM-5.2 fully opens, supports 1M context and will be open-sourced next week
-
Hermes Agent Desktop released with one-click model switching via SiliconFlow
-
From 0 to 1: a quick-start guide to WorkBuddy, a general-purpose agent product for China
-
MiMo Code Released and Open-Sourced: Model-Agent Collaborative Optimization, Toward a Self-Evolving Era
-
Xiaomi releases and open-sources terminal AI coding assistant MiMo Code V0.1.0 under MIT license
-
Moore Threads open-sources MusaCoder code LLM, 9B/27B parameters trained end-to-end on domestic GPUs
-
Zhipu proposes A-share offering and STAR Market listing
-
Zhipu GLM-5.1 High-Speed released, setting a new global record for LLM API speed
-
GLM-5.1 high-speed version released: 400 tokens/s, flagship capability at the world's fastest API speed
-
Zhipu launches GLM-5.1 high-speed version
-
Zhipu, Yuxun Network and Tsinghua propose ZCube network architecture to break large-model inference network bottlenecks
-
Integrating multiple AI models for development in VS Code
-
GLM-5.1 ranks first among open-source models on AA's new Coding Agent benchmark
Monthly archive
Frequently asked questions
- What is Zhipu AI?
- Zhipu AI builds the GLM model family and enterprise AI services in China.
- What is the latest Zhipu AI news?
- The most recent story is “NVIDIA NIM free DeepSeek / GLM Flash APIs: get a Key; Kimi K3.1 Pro is Playground-only for now” (November 7, 2026). This page lists Zhipu AI coverage in reverse chronological order so you can keep following it.
- How can I follow Zhipu AI news?
- Open this page for the latest Zhipu AI updates (69 stories). You can also subscribe to the site-wide RSS or JSON Feed; pages refresh about every 30 minutes.