Gate News message, April 2, Zhipu AI released the multimodal coding foundation model GLM-5V-Turbo, designed specifically for visual programming scenarios. The model natively supports multimodal input understanding such as images, video, design drafts, and document layouts, and can also call multimodal tools such as framing, screenshots, and reading webpages, with the context window expanded to 200k.

Zhipu released GLM-5.1, making it available to all GLM Coding Plan users. In the coding ability test, GLM-5.1 scored 45.3, significantly better than the previous generation’s 35.4, and close to Claude Opus 4.6, showing clear progress.

SWE-rebench is a real-time software engineering task benchmark. The latest leaderboard update removed demo and operation restrictions, and added auxiliary evaluation tasks. Claude Opus 4.6 ranks first with 65.3%, GLM-5 ranks third with 62.8%, and four Chinese models made the top ten, demonstrating progress in Chinese models.

On March 23, SWE-rebench published its latest benchmark rankings, with Claude Opus 4.6 topping the list. This update removed example demonstrations and step limitations, while adding auxiliary evaluation tasks. GLM-5 ranked third among open-source models, with Chinese models collectively occupying four positions.

Zhipu Z.ai's global head Li Zixuan revealed that GLM-5 leads human benchmarks in the Prediction Arena prediction market with a 42.99% return rate, becoming the only AI model to surpass humans. The model executed 98 trades with outstanding performance.

Gate News, on March 21, Zhipu Z.ai global head Li Zixuan posted on the X platform stating: "Don't panic. GLM-5.1 will be open source." (Don't panic, GLM-5.1 will be open source.) The release date and specific details of the model have not yet been disclosed.

Zhipu AI released the GLM-5-Turbo model on March 16th as a high-speed optimized version of GLM-5, specifically customized for OpenClaw agent scenarios. The model delivers 2-3x speed improvements, supports 200K context and 128K maximum output, and has optimized tool calling stability. It demonstrated significantly superior performance compared to GLM-5 in evaluations, alongside corresponding increases to API pricing and the introduction of personal user packages.

Tencent Cloud announces that starting March 13, the billing strategy for some models on the intelligent agent development platform will be adjusted. The HY2.0 series models will see a significant price increase, while models such as GLM 5, MiniMax 2.5, and Kimi 2.5 will end their free public testing and transition to paid plans. Package users can deduct part of the costs.

PANews February 25 News, according to Zhitong Caijing, Alibaba Cloud BaiLian has launched the Coding Plan, which includes API services for four major open-source models: Qwen3.5, GLM-5, MiniMax M2.5, and Kimi K2.5. Currently, among global cloud providers, only Alibaba Cloud offers this service.

Zhipu issues an apology for the release of its AI programming subscription plan, the GLM Coding Plan, citing issues such as insufficient transparency of rules during the revision, slow rollout of the GLM-5 grayscale phase, and an imperfect upgrade mechanism for existing users. The company also proposes solutions and compensation measures, including allowing Lite and Pro users to request refunds.