Z.ai: upcoming dates

Recent

• AI10/12/2026@TechDesk

  GLM-5.3-Flash Released

  Yard 1, Zhongguancun East Road, Haidian District, Beijing, ChinaZ.ai

  

  Starts 08/26/2026CONFIRMEDDate confidence: CONFIRMEDStated as firmHow firmly the source words the date: confirmed, scheduled, estimated, delayed or unverified.Show all CONFIRMED pins86% CONFIDENCEOverall confidence: 86%How well the pin's source and references back up its dates.Weighted average of how firmly 6 references, the source included, support the pin's start and end times; a reference counts half as much for every 180 days older than the newestShow all pins at 75% confidence or better54 days ago

  Z.ai's developer release notes log GLM-5.3-Flash on 26 August 2026, INSIDE^[4] reports Z.ai's own X post that day naming it as the model that had circulated anonymously as "Ox Alpha" from 20 August, with weights published the same day, and DataNorth^[5] says it was "released on 26 August 2026". Epoch AI lists 20 August, which is the anonymous OpenRouter debut, not the maker's release.

  Z.ai releases GLM-5.3-Flash, a 320B-parameter (18B active) natively multimodal open-weight model on a newly trained base that it says beats GLM-5.2 at one-tenth of the price, listed at $0.15 per million input tokens.

  120
• AI10/12/2026@TechDesk

  GLM-5.3 Released

  Yard 1, Zhongguancun East Road, Haidian District, Beijing, ChinaZ.ai

  

  Starts 08/14/2026CONFIRMEDDate confidence: CONFIRMEDStated as firmHow firmly the source words the date: confirmed, scheduled, estimated, delayed or unverified.Show all CONFIRMED pins86% CONFIDENCEOverall confidence: 86%How well the pin's source and references back up its dates.Weighted average of how firmly 6 references, the source included, support the pin's start and end times; a reference counts half as much for every 180 days older than the newestShow all pins at 75% confidence or better66 days ago

  Z.ai put GLM-5.3 into its API, GLM Coding Plan and ZCode client on 14 August 2026: Reuters (via The Standard^[6]) is dated 14 August and says Z.ai "said on Friday", and DataNorth^[5] says it was "released on 14 August 2026". Z.ai's developer release notes log the entry under 18 August and the open weights followed on 28 August; both are later steps, not the release.

  Z.ai (Zhipu AI) releases GLM-5.3, a post-training-only upgrade of GLM-5.2 that scores 28.3 on Terminal Bench 3.0 and 84.5% on the CyberGym vulnerability benchmark, with a 1M-token context and open weights two weeks later.

  30
• AI10/12/2026@TechDesk

  GLM-5.2 Released

  Yard 1, Zhongguancun East Road, Haidian District, Beijing, ChinaZ.ai

  

  Starts 06/16/2026CONFIRMEDDate confidence: CONFIRMEDStated as firmHow firmly the source words the date: confirmed, scheduled, estimated, delayed or unverified.Show all CONFIRMED pins84% CONFIDENCEOverall confidence: 84%How well the pin's source and references back up its dates.Weighted average of how firmly 6 references, the source included, support the pin's start and end times; a reference counts half as much for every 180 days older than the newestShow all pins at 75% confidence or better125 days ago

  Z.ai's Hugging Face^[6] repository, API and developer release notes all date GLM-5.2 to 16 June 2026 and Epoch^[5] AI lists "June 16, 2026". Coding Plan subscribers had it from 13 June (a Saturday); AI Weekly^[2] says that launch promised the open weights and API for 16 June, so 13 June was a staged first access, not the full release.

  Z.ai (Zhipu AI) releases GLM-5.2, an MIT-licensed 753B-parameter flagship built for long-horizon coding with a 1M-token context window, scoring 62.1 on SWE-bench Pro and 81.0 on Terminal Bench 2.1.

  40
• AI10/12/2026@TechDesk

  GLM-5.1 Released

  Yard 1, Zhongguancun East Road, Haidian District, Beijing, ChinaZ.ai

  

  Starts 04/07/2026CONFIRMEDDate confidence: CONFIRMEDStated as firmHow firmly the source words the date: confirmed, scheduled, estimated, delayed or unverified.Show all CONFIRMED pins85% CONFIDENCEOverall confidence: 85%How well the pin's source and references back up its dates.Weighted average of how firmly 6 references, the source included, support the pin's start and end times; a reference counts half as much for every 180 days older than the newestShow all pins at 75% confidence or better195 days ago

  VentureBeat^[4] (7 April 2026) says Z.ai "has unveiled GLM-5.1 today under a permissive MIT License"; Z.ai's developer release notes and Epoch^[6] AI both list 7 April 2026. The Hugging Face^[5] repository was created on 3 April and weights were uploaded from 4 April, before the public announcement.

  Z.ai (Zhipu AI) releases GLM-5.1 as MIT-licensed open weights, a 754B-parameter model built to work autonomously on one task for up to 8 hours, topping SWE-Bench Pro at 58.4.

  110
• AI10/12/2026@TechDesk

  GLM-5V-Turbo Released

  Yard 1, Zhongguancun East Road, Haidian District, Beijing, ChinaZ.ai

  

  Starts 04/01/2026CONFIRMEDDate confidence: CONFIRMEDStated as firmHow firmly the source words the date: confirmed, scheduled, estimated, delayed or unverified.Show all CONFIRMED pins84% CONFIDENCEOverall confidence: 84%How well the pin's source and references back up its dates.Weighted average of how firmly 5 references, the source included, support the pin's start and end times; a reference counts half as much for every 180 days older than the newestShow all pins at 75% confidence or better201 days ago

  Vercel's^[5] AI Gateway describes GLM-5V-Turbo as "released April 1, 2026" and OpenRouter lists z-ai/glm-5v-turbo with a 1 April 2026 creation date; SiliconFlow's^[4] 2 April post says it is "now officially live" and The Decoder^[2] reported it on 3 April. Z.ai's developer docs carry no dated announcement for this model, so the day rests on those listings.

  Z.ai (Zhipu AI) releases GLM-5V-Turbo, its first multimodal coding foundation model, which turns design mockups, screenshots and video into runnable front-end code and scores 94.8 on Design2Code against Claude Opus 4.6's 77.3.

  110
• AI10/12/2026@TechDesk

  GLM-5 Released

  Yard 1, Zhongguancun East Road, Haidian District, Beijing, ChinaZ.ai

  

  Starts 02/11/2026CONFIRMEDDate confidence: CONFIRMEDStated as firmHow firmly the source words the date: confirmed, scheduled, estimated, delayed or unverified.Show all CONFIRMED pins85% CONFIDENCEOverall confidence: 85%How well the pin's source and references back up its dates.Weighted average of how firmly 6 references, the source included, support the pin's start and end times; a reference counts half as much for every 180 days older than the newestShow all pins at 75% confidence or better250 days ago

  Epoch^[5] AI lists GLM-5 as released 11 February 2026 and Maxime Labonne's Hugging Face^[3][6] write-up says "On February 11th, 2026 ... Z.ai officially released GLM-5"; the Hugging Face weights were uploaded from 06:15 UTC that day. The Decoder^[4] dates its report 12 February and Z.ai's release notes list 2026-02-12, which is the same announcement on Beijing time.

  Z.ai (Zhipu AI) releases GLM-5, a 744B-parameter (40B active) MIT-licensed model for long-horizon agentic engineering that scores 77.8 on SWE-bench Verified and became the first open-weight model to reach 50 on the Artificial Analysis index.

  120
• AI10/12/2026@TechDesk

  GLM-4.7-Flash Released

  Yard 1, Zhongguancun East Road, Haidian District, Beijing, ChinaZ.ai

  

  Starts 01/19/2026CONFIRMEDDate confidence: CONFIRMEDStated as firmHow firmly the source words the date: confirmed, scheduled, estimated, delayed or unverified.Show all CONFIRMED pins85% CONFIDENCEOverall confidence: 85%How well the pin's source and references back up its dates.Weighted average of how firmly 5 references, the source included, support the pin's start and end times; a reference counts half as much for every 180 days older than the newestShow all pins at 75% confidence or better273 days ago

  Gigazine^[5] (20 January 2026 JST) says Z.ai "released a lightweight AI model called GLM-4.7-Flash on January 19, 2026" and embeds Z.ai's X posts dated January 19, 2026; the Hugging Face^[1] repository was created and populated on 19 January. Some trackers show 20 January, a time-zone effect.

  Z.ai releases GLM-4.7-Flash, a 30B-A3B mixture-of-experts open-weight model that beats gpt-oss-20b and Qwen3-30B-A3B-Thinking on most coding and agent benchmarks, with a free API tier.

  120
• AI10/12/2026@TechDesk

  GLM-4.7 Released

  Yard 1, Zhongguancun East Road, Haidian District, Beijing, ChinaZ.ai

  

  Starts 12/22/2025CONFIRMEDDate confidence: CONFIRMEDStated as firmHow firmly the source words the date: confirmed, scheduled, estimated, delayed or unverified.Show all CONFIRMED pins87% CONFIDENCEOverall confidence: 87%How well the pin's source and references back up its dates.Weighted average of how firmly 6 references, the source included, support the pin's start and end times; a reference counts half as much for every 180 days older than the newestShow all pins at 75% confidence or better301 days ago

  Gigazine^[4] (23 December 2025) says Z.ai "released the coding-specialized AI model GLM-4.7 on December 22, 2025", Epoch^[5] AI lists 22 December 2025 and Z.ai's developer release notes carry the same date; the Hugging Face^[6] repository was created and populated on 22 December.

  Z.ai (Zhipu AI) releases GLM-4.7, an MIT-licensed coding-focused flagship that scores 73.8 on SWE-bench Verified and 87.4 on τ²-Bench, ranking first among open models on Code Arena at the end of 2025.

  110
• AI10/12/2026@TechDesk

  GLM-4.6V and GLM-4.6V-Flash Released

  Yard 1, Zhongguancun East Road, Haidian District, Beijing, ChinaZ.ai

  

  Starts 12/08/2025CONFIRMEDDate confidence: CONFIRMEDStated as firmHow firmly the source words the date: confirmed, scheduled, estimated, delayed or unverified.Show all CONFIRMED pins87% CONFIDENCEOverall confidence: 87%How well the pin's source and references back up its dates.Weighted average of how firmly 5 references, the source included, support the pin's start and end times; a reference counts half as much for every 180 days older than the newestShow all pins at 75% confidence or better315 days ago

  VentureBeat^[4] reported on 8 December 2025 (5:03 pm PT) that Z.ai "has released its GLM-4.6V series" and OpenRouter^[5] lists z-ai/glm-4.6v on 8 December; the Hugging Face^[1] repositories were uploaded on 7 December UTC and docs.z.ai^[2]'s launch case-study videos are labelled 1208.

  Z.ai releases GLM-4.6V, a 106B vision-language mixture-of-experts model, and the 9B GLM-4.6V-Flash under the MIT license, the first GLM-V models with native function calling, trained for a 128K-token context and priced at $0.30 per million input tokens.

  110
• AI10/12/2026@TechDesk

  GLM-4.6 Released

  Yard 1, Zhongguancun East Road, Haidian District, Beijing, ChinaZ.ai

  

  Starts 09/30/2025CONFIRMEDDate confidence: CONFIRMEDStated as firmHow firmly the source words the date: confirmed, scheduled, estimated, delayed or unverified.Show all CONFIRMED pins84% CONFIDENCEOverall confidence: 84%How well the pin's source and references back up its dates.Weighted average of how firmly 6 references, the source included, support the pin's start and end times; a reference counts half as much for every 180 days older than the newestShow all pins at 75% confidence or better1.1 years ago

  Z.ai's developer release notes list "2025-09-30 GLM-4.6" ("We've launched GLM-4.6, the flagship coding model"); Wikipedia^[2] cites Z.ai's launch post dated 30 September 2025, the Hugging Face^[6] weights were uploaded late on 29 September UTC (early 30 September in Beijing) and The Decoder^[4] reported it on 1 October.

  Z.ai (Zhipu AI) releases GLM-4.6, its flagship coding model with a 200K-token context window, which wins 48.6% of head-to-head real-world coding tasks against Claude Sonnet 4 and uses over 30% fewer tokens than GLM-4.5.

  00
• AI10/12/2026@TechDesk

  GLM-4.5V Released

  Yard 1, Zhongguancun East Road, Haidian District, Beijing, ChinaZ.ai

  

  Starts 08/11/2025CONFIRMEDDate confidence: CONFIRMEDStated as firmHow firmly the source words the date: confirmed, scheduled, estimated, delayed or unverified.Show all CONFIRMED pins82% CONFIDENCEOverall confidence: 82%How well the pin's source and references back up its dates.Weighted average of how firmly 6 references, the source included, support the pin's start and end times; a reference counts half as much for every 180 days older than the newestShow all pins at 75% confidence or better1.2 years ago

  The model card's project updates read "2025/08/11: We released GLM-4.5V"; OpenRouter^[6] lists z-ai/glm-4.5v on 11 August 2025 and the arXiv^[5] paper's GLM-4.5V revision followed on 13 August, while the Hugging Face^[1] weights were uploaded on 10 August UTC.

  Z.ai releases GLM-4.5V, an open-weight 106B vision-language mixture-of-experts model with 12B active parameters, built on GLM-4.5-Air with a Thinking Mode switch and GUI-agent and grounding skills, and claims state-of-the-art results among similar-size open models on 42 public benchmarks.

  170
• AI10/12/2026@TechDesk

  GLM-4.5 and GLM-4.5-Air Released

  Yard 1, Zhongguancun East Road, Haidian District, Beijing, ChinaZ.ai

  

  Starts 07/28/2025CONFIRMEDDate confidence: CONFIRMEDStated as firmHow firmly the source words the date: confirmed, scheduled, estimated, delayed or unverified.Show all CONFIRMED pins83% CONFIDENCEOverall confidence: 83%How well the pin's source and references back up its dates.Weighted average of how firmly 6 references, the source included, support the pin's start and end times; a reference counts half as much for every 180 days older than the newestShow all pins at 75% confidence or better1.2 years ago

  Z.ai's GLM-4.5 launch post is dated 28 July 2025 (as cited by Wikipedia^[2]), and the weights for both GLM-4.5 and GLM-4.5-Air were uploaded to Hugging Face^[6] that day (04:44 UTC and 09:14 UTC). The repositories were created earlier, on 20 July, and Epoch AI lists 3 August and 20 July respectively, which are not the public release.

  Zhipu AI (Z.ai) releases GLM-4.5 (355B, 32B active) and GLM-4.5-Air (106B, 12B active), MIT-licensed hybrid-reasoning models for agents that rank third and sixth across 12 benchmarks, with API prices as low as $0.20 per million input tokens.

  80