Meta distilled the dense 30B model Muse Glimmer from the larger Muse Spark, training it on synthetic agent trajectories in simulated applications. Its 4-bit weights are under 20 GB, allowing it to run on a single RTX 5090. According to a report dated October 22, 2026, on the multi-tool benchmark MCPAtlas, Muse Glimmer scored 75.5, higher than Gemma4-31B's 54.2 and Qwen3.6-27B's 62.5; however, on OSWorld-Verified it lagged behind Qwen3.6-27B's 75.6 with a score of 65.9. There are currently no subsequent developments.
Meta announced on October 22, 2026, the issuance of a new class of shares to finance its AI plans. At the earnings call the same day, it was disclosed that within six weeks of launch, more than 10 million people in the United States and Canada had set up the AI product Muse, with about 25% of users using it daily; shopping is the most common use case, with an average order value exceeding $200; the company plans to bring Muse to Meta Glasses in November 2026 and to launch the hardware device Muse Charm. Chief Financial Officer Susan Li said that, apart from subscriptions, Muse has not yet been monetized. There are currently no subsequent developments.
Meta announced on October 22, 2026, the issuance of a new class of shares to finance its AI plans. At the earnings call the same day, it was disclosed that within six weeks of launch, more than 10 million people in the United States and Canada had set up the AI product Muse, with about 25% of users using it daily; shopping is the most common use case, with an average order value exceeding $200; the company plans to bring Muse to Meta Glasses in November 2026 and to launch the hardware device Mus
Meta announced on October 22, 2026, the issuance of a new class of shares to finance its AI plans. At the earnings call the same day, it was disclosed that within six weeks of launch, more than 10 million people in the United States and Canada had set up the AI product Muse, with about 25% of users using it daily; shopping is the most common use case, with an average order value exceeding $200; the company plans to bring Muse to Meta Glasses in November 2026 and to launch the hardware device Muse Charm. Chief Financial Officer Susan Li said that, apart from subscriptions, Muse has not yet been monetized. There are currently no subsequent developments.
OpenCode announced that Meta Muse Wonder is now available on its platform. The news was released on October 22, 2026; currently this is the only factual evidence, with no subsequent developments or official supplementary statements yet.
OpenCode announced that Meta Muse Wonder is now available on its platform. The news was released on October 22, 2026; currently this is the only factual evidence, with no subsequent developments or official supplementary statements yet.
Rohan Paul offered the assessment that Meta's AI agent Muse reached 10.2 million downloads in the US within 40 days of launch. The data he cited also shows that historically only 11 apps have reached this pace, most of them mobile games. On retention, Muse's US 7-day retention rate is 30% and its 14-day retention rate is 23%, compared with competitor averages of 19% and 16%, respectively. On October 16, its daily active users reached 2.5 million, up 15% in a week, but still less than a quarter of its US downloads. At present, this assessment is seen only from Rohan Paul as a single source, with no other independent sources or official evidence to cross-verify it.
Rohan Paul offered the assessment that Meta's AI agent Muse reached 10.2 million downloads in the US within 40 days of launch. The data he cited also shows that historically only 11 apps have reached this pace, most of them mobile games. On retention, Muse's US 7-day retention rate is 30% and its 14-day retention rate is 23%, compared with competitor averages of 19% and 16%, respectively. On October 16, its daily active users reached 2.5 million, up 15% in a week, but still less than a quarter o
Rohan Paul offered the assessment that Meta's AI agent Muse reached 10.2 million downloads in the US within 40 days of launch. The data he cited also shows that historically only 11 apps have reached this pace, most of them mobile games. On retention, Muse's US 7-day retention rate is 30% and its 14-day retention rate is 23%, compared with competitor averages of 19% and 16%, respectively. On October 16, its daily active users reached 2.5 million, up 15% in a week, but still less than a quarter of its US downloads. At present, this assessment is seen only from Rohan Paul as a single source, with no other independent sources or official evidence to cross-verify it.
On October 21, 2026, Meta published a paper pointing out that the failure point of proactive assistants (such as AI glasses) lies in autonomously judging when to speak, not just what to say. The team built the ProPerMem benchmark, using first-person video paired with interference-rich full-year memories averaging 618K tokens, and featuring manually verified trigger windows. The paper reports that all baselines scored at most 15%, even frontier models using retrieval-augmented memory; when oracle evidence was provided within the window, GPT-5.3-Chat reached 96%, but when it was left to choose the timing itself, this dropped to 38%, and after adding retrieval it fell to 15%. Grasping both content and timing while reducing trigger frequency can raise accuracy from under 10% to 42.4%. The work is currently released as a paper, with no subsequent progress yet.
On October 21, 2026, Meta published a paper pointing out that the failure point of proactive assistants (such as AI glasses) lies in autonomously judging when to speak, not just what to say. The team built the ProPerMem benchmark, using first-person video paired with interference-rich full-year memories averaging 618K tokens, and featuring manually verified trigger windows. The paper reports that all baselines scored at most 15%, even frontier models using retrieval-augmented memory; when oracle
Alibaba Cloud OSS launches the new version of MetaQuery. According to an article published by the Alibaba Cloud Developer Community AI on October 21, 2026, Alibaba Cloud OSS has launched a "object bucket + vector bucket + table bucket" three-bucket-in-one architecture, unifying unstructured data, vector data, and structured data onto the same storage foundation. The article states that a single file in an object bucket supports up to 48.8TB, with 12 nines of data reliability; the vector bucket supports trillion-scale vector volumes and millisecond-level similarity retrieval; the table bucket natively integrates Apache Iceberg, and concurrent write TPS for 500 tables can reach more than 10 times that of self-built Iceberg solutions. The article also breaks down the four-layer cloud-native architecture and hands-on steps such as bucket creation, SDK uploads, and batch operations and maintenance with ossutil. There are currently no subsequent development facts.
Alibaba Cloud OSS launches the new version of MetaQuery. According to an article published by the Alibaba Cloud Developer Community AI on October 21, 2026, Alibaba Cloud OSS has launched a "object bucket + vector bucket + table bucket" three-bucket-in-one architecture, unifying unstructured data, vector data, and structured data onto the same storage foundation. The article states that a single file in an object bucket supports up to 48.8TB, with 12 nines of data reliability; the vector bucket s
Alibaba Cloud OSS launches the new version of MetaQuery. According to an article published by the Alibaba Cloud Developer Community AI on October 21, 2026, Alibaba Cloud OSS has launched a "object bucket + vector bucket + table bucket" three-bucket-in-one architecture, unifying unstructured data, vector data, and structured data onto the same storage foundation. The article states that a single file in an object bucket supports up to 48.8TB, with 12 nines of data reliability; the vector bucket supports trillion-scale vector volumes and millisecond-level similarity retrieval; the table bucket natively integrates Apache Iceberg, and concurrent write TPS for 500 tables can reach more than 10 times that of self-built Iceberg solutions. The article also breaks down the four-layer cloud-native architecture and hands-on steps such as bucket creation, SDK uploads, and batch operations and maintenance with ossutil. There are currently no subsequent development facts.
US Judge Thomas Hixson ordered Meta to provide the command history of the download servers and to list all personnel involved in approving the downloading of pirated ebook libraries via BitTorrent. Meta previously said it could not identify the specific decision-makers, that such decisions were usually team consensus, and that downloading torrents was a technical decision by the executing employees. The book publishers and authors suing Meta want to find out who ordered the downloading of pirated ebooks from Anna's Archive and LibGen for training AI models.
US Judge Thomas Hixson ordered Meta to provide the command history of the download servers and to list all personnel involved in approving the downloading of pirated ebook libraries via BitTorrent. Meta previously said it could not identify the specific decision-makers, that such decisions were usually team consensus, and that downloading torrents was a technical decision by the executing employees. The book publishers and authors suing Meta want to find out who ordered the downloading of pirate
US Judge Thomas Hixson ordered Meta to provide the command history of the download servers and to list all personnel involved in approving the downloading of pirated ebook libraries via BitTorrent. Meta previously said it could not identify the specific decision-makers, that such decisions were usually team consensus, and that downloading torrents was a technical decision by the executing employees. The book publishers and authors suing Meta want to find out who ordered the downloading of pirated ebooks from Anna's Archive and LibGen for training AI models.
Meta is developing Confidential VMs functionality on the Muse web client, allowing users to run Muse tasks in an encrypted confidential environment. Meta previously previewed that this feature would launch before the end of 2026. A newly surfaced hidden setting shows that the feature will initially be available only to Meta AI subscribers; after toggling it on, voice mode, image and video generation, and wearable device support will be temporarily unavailable. Meta says conversations run in a secure encrypted environment and that Muse data will not be used to improve Meta's AI. The feature is currently still in development, and there is no further evidence of an official launch date.
Meta is developing Confidential VMs functionality on the Muse web client, allowing users to run Muse tasks in an encrypted confidential environment. Meta previously previewed that this feature would launch before the end of 2026. A newly surfaced hidden setting shows that the feature will initially be available only to Meta AI subscribers; after toggling it on, voice mode, image and video generation, and wearable device support will be temporarily unavailable. Meta says conversations run in a se
On October 21, 2026, Meta published a paper pointing out that the failure point of proactive assistants (such as AI glasses) lies in autonomously judging when to speak, not just what to say. The team built the ProPerMem benchmark, using first-person video paired with interference-rich full-year memories averaging 618K tokens, and featuring manually verified trigger windows. The paper reports that all baselines scored at most 15%, even frontier models using retrieval-augmented memory; when oracle evidence was provided within the window, GPT-5.3-Chat reached 96%, but when it was left to choose the timing itself, this dropped to 38%, and after adding retrieval it fell to 15%. Grasping both content and timing while reducing trigger frequency can raise accuracy from under 10% to 42.4%. The work is currently released as a paper, with no subsequent progress yet.