Search “Google”

Google releases world model Genie 4

On October 27, 2026, Google DeepMind released the world model Genie 4, focused on real-time interactive simulation and introducing native audio to the series for the first time. According to the official introduction, the model is built on capabilities from the Gemini family: Gemini Omni is responsible for making the simulation match the appearance and behavior of the real world, while other Gemini models provide the complex reasoning needed to maintain the simulation's coherence; users can control scenes in real time using natural language, and each world also instantly generates ambient sound, music, and interactive dialogue. The next day, Google DeepMind further explained that Gemini 3.8 Flash breaks down goals in a loop and checks frames to maintain causal coherence across multi-step events, and that Gemini 4 Argon has also been used to jointly simulate tasks with Gemini Robotics; the company said it remains an early experimental model, that object permanence and complex physical interactions still drift during long runs, that all generated content carries a SynthID watermark, and that it is currently being previewed to trusted testers first. On October 28, Ethan Mollick posted that Genie 4 can run DOOM, and tested it generating a game-like environment as it goes, saying the experience is not perfect. On October 29, Peter Yang used Genie 4 to build 5 worlds in real time, arguing that world models are the next big direction for AI after LLMs.

Gemini Notebook will launch Connectors, initially supporting only Google Drive

Testing Catalog reports that Gemini Notebook is about to gain the Connectors feature, but initially only one connector, Google Drive, will be available. Users can search Google Drive files directly in chat and can disable the feature as needed; officials say the connector only retrieves workspace documents, and original files will not be modified. More Connectors are expected to be added later. There is currently no further development progress.

View story:Gemini Notebook to Launch Connectors, Initially Supporting Only Google Drive

Google DeepMind paper proposes CALLIOPE topic tree method

Google DeepMind published a paper proposing the CALLIOPE topic tree construction method: have a large language model tag each document with a small number of keywords and add noise, keep only the high-frequency keywords, and then have a large language model propose topics and classify documents based on them, thereby generating a clearer topic map under strict privacy constraints. Under the strictest privacy setting, human raters preferred CALLIOPE over the embedding-vector-based URANIA on all 3 datasets, with the highest ratio being 76% versus 18%. This method consumes 2 to 5 times as many tokens as URANIA. There are currently no follow-up developments.

Gemini Notebook to Launch Connectors, Initially Supporting Only Google Drive

Testing Catalog reports that Gemini Notebook is about to gain the Connectors feature, but initially only one connector, Google Drive, will be available. Users can search Google Drive files directly in chat and can disable the feature as needed; officials say the connector only retrieves workspace documents, and original files will not be modified. More Connectors are expected to be added later. There is currently no further development progress.

Google paper proposes engineering task benchmark PLE

Google has published a new paper proposing the PLE benchmark for entity engineering tasks. The benchmark contains 85 electrical, civil, structural, and mechanical engineering tasks written or verified by practicing engineers. According to a report by Rohan Paul, the strongest agent scores only 50.6% on PLE: Claude Opus 5.5 in Claude Code ranks first among 14 models at 50.6%, consuming 27M tokens and 63 minutes per task, requiring reading about 61 project files and delivering schedules, reports, or 2D/3D drawings. The report also states that in design tasks its output satisfies 91% of constraints, but only 41.3% satisfy all hard constraints. There are currently no follow-up developments.

TestingCatalog predicts that this year Google or another company will still release a large model.

When commenting on Zachary Nado's statement that "Gemini [redacted] hasn't arrived yet," TestingCatalog predicted: this year we will either still be waiting for a large model from Google, or [redacted]. This prediction was published by Testing Catalog on October 29, 2026, and there has been no subsequent development or official response so far.

TestingCatalog predicts that this year Google or another company will also release a large model

When commenting on Zachary Nado's statement that "Gemini [redacted] hasn't arrived yet," TestingCatalog predicted: this year we will either still be waiting for a large model from Google, or [redacted]. This prediction was published by Testing Catalog on October 29, 2026, and there has been no subsequent development or official response so far.

View story:TestingCatalog predicts that this year Google or another company will still release a large model.

Google Releases EmbeddingGemma 2 Paper

Google has released the EmbeddingGemma 2 paper, introducing an Apache 2.0-licensed on-device multimodal embedding model that maps text, code, images, audio, and video into the same vector space. Built on Gemma 4, the model learns from the larger Gemini Embedding 2 and is tuned with contrastive training; with 744M parameters, it outperforms models 2 to 12 times larger on multiple tasks. On the Pixel 11 Pro, the quantized text weights occupy about 191MB of active memory, while the full multimodal model takes about 567MB; the context window is expanded 4 times to 8K tokens, accommodating about 5.5 minutes of audio, 29 images, or 58 video frames. Embedding speech directly scores 69.54 on the MSEB spoken query retrieval benchmark, higher than the 57.59 achieved by transcribing first and then retrieving, and code retrieval on MTEB Code improves from 68.76 in EmbeddingGemma 1 to 78.68. There are currently no further developments.

Google DeepMind paper: Training models to write moral arguments makes them more decisive but does not improve their judgment

Google DeepMind published a paper investigating the effects of training Gemma-4 4B with reinforcement learning to write moral arguments. The results showed that on ambiguous cases in MoralChoice, the proportion of answers that changed due to swapping the order of options dropped from 14.3% to 6.6%, but the proportion of sycophantic agreement with users only went from 45.3% to 44.0%, and accuracy on moral fable questions actually fell from 56.5% to 53.4%. Based on this, the paper recommends that when evaluating advice- or judgment-type tasks, consistency and correctness should be tested separately.

Google DeepMind paper: Training models to write moral arguments makes them more decisive, but their judgments do not improve

Google DeepMind published a paper investigating the effects of training Gemma-4 4B with reinforcement learning to write moral arguments. The results showed that on ambiguous cases in MoralChoice, the proportion of answers that changed due to swapping the order of options dropped from 14.3% to 6.6%, but the proportion of sycophantic agreement with users only went from 45.3% to 44.0%, and accuracy on moral fable questions actually fell from 56.5% to 53.4%. Based on this, the paper recommends that when evaluating advice- or judgment-type tasks, consistency and correctness should be tested separately.

View story:Google DeepMind paper: Training models to write moral arguments makes them more decisive but does not improve their judgment

Kimi integrates with Gmail, Google Drive, and Google Calendar

English text is being prepared.

Yuchen Jin raises the question of whether we are entering a new round of the "Google comeback" cycle

On October 29, 2026, Yuchen Jin posted a question: whether Google has once again entered a new "Google is back" comeback cycle. He said that Google's new model, Gemini 4 Carbon, surpasses Opus 5.5 and GPT-6 Astra on most benchmarks and ranks first on the Artificial Analysis Intelligence Index v5. He also said it remains to be seen whether this performance can shake Anthropic's IPO. At present, this judgment remains at the level of Yuchen Jin's personal question and observation, with no subsequent factual developments yet.

Yuchen Jin raises the question of whether we are entering a new "Google comeback" cycle

On October 29, 2026, Yuchen Jin posted a question: whether Google has once again entered a new "Google is back" comeback cycle. He said that Google's new model, Gemini 4 Carbon, surpasses Opus 5.5 and GPT-6 Astra on most benchmarks and ranks first on the Artificial Analysis Intelligence Index v5. He also said it remains to be seen whether this performance can shake Anthropic's IPO. At present, this judgment remains at the level of Yuchen Jin's personal question and observation, with no subsequent factual developments yet.

View story:Yuchen Jin raises the question of whether we are entering a new round of the "Google comeback" cycle

Google Antigravity adds a plugin marketplace

Google Antigravity adds a plugin marketplace

TestingCatalog says Google has terminated the Jules AI agent

On October 30, 2026, TestingCatalog reported that Google will terminate its AI coding agent Jules. Phoenix Technology reported the same day that Google will gradually shut down Jules, which was launched in December 2024, and will integrate the experience accumulated from the tool into its AI coding platform Antigravity; Jules' official X account stated that Jules was originally an experiment exploring how asynchronous coding agents could help developers ship faster, and that its experience has already influenced multiple Google products. cnBeta reported the same day that Jules will gradually cease operations, less than two years after its launch, and that the team is integrating the experience into Antigravity so that users can both autonomously execute background tasks in remote environments and use an agent-centric development environment. The current status is gradual shutdown and experience integration.

TestingCatalog says Google terminates Jules AI agent

On October 30, 2026, TestingCatalog reported that Google will terminate its AI coding agent Jules. Phoenix Technology reported the same day that Google will gradually shut down Jules, which was launched in December 2024, and will integrate the experience accumulated from the tool into its AI coding platform Antigravity; Jules' official X account stated that Jules was originally an experiment exploring how asynchronous coding agents could help developers ship faster, and that its experience has already influenced multiple Google products. cnBeta reported the same day that Jules will gradually cease operations, less than two years after its launch, and that the team is integrating the experience into Antigravity so that users can both autonomously execute background tasks in remote environments and use an agent-centric development environment. The current status is gradual shutdown and experience integration.

View story:TestingCatalog says Google has terminated the Jules AI agent

Google DeepMind paper proposes CALLIOPE topic tree method

Google DeepMind published a paper proposing the CALLIOPE topic tree construction method: have a large language model tag each document with a small number of keywords and add noise, keep only the high-frequency keywords, and then have a large language model propose topics and classify documents based on them, thereby generating a clearer topic map under strict privacy constraints. Under the strictest privacy setting, human raters preferred CALLIOPE over the embedding-vector-based URANIA on all 3 datasets, with the highest ratio being 76% versus 18%. This method consumes 2 to 5 times as many tokens as URANIA. There are currently no follow-up developments.

View story:Google DeepMind paper proposes CALLIOPE topic tree method

Google paper proposes engineering task benchmark PLE

Google has published a new paper proposing the PLE benchmark for entity engineering tasks. The benchmark contains 85 electrical, civil, structural, and mechanical engineering tasks written or verified by practicing engineers. According to a report by Rohan Paul, the strongest agent scores only 50.6% on PLE: Claude Opus 5.5 in Claude Code ranks first among 14 models at 50.6%, consuming 27M tokens and 63 minutes per task, requiring reading about 61 project files and delivering schedules, reports, or 2D/3D drawings. The report also states that in design tasks its output satisfies 91% of constraints, but only 41.3% satisfy all hard constraints. There are currently no follow-up developments.

View story:Google paper proposes engineering task benchmark PLE

Google releases EmbeddingGemma 2 paper

Google has released the EmbeddingGemma 2 paper, introducing an Apache 2.0-licensed on-device multimodal embedding model that maps text, code, images, audio, and video into the same vector space. Built on Gemma 4, the model learns from the larger Gemini Embedding 2 and is tuned with contrastive training; with 744M parameters, it outperforms models 2 to 12 times larger on multiple tasks. On the Pixel 11 Pro, the quantized text weights occupy about 191MB of active memory, while the full multimodal model takes about 567MB; the context window is expanded 4 times to 8K tokens, accommodating about 5.5 minutes of audio, 29 images, or 58 video frames. Embedding speech directly scores 69.54 on the MSEB spoken query retrieval benchmark, higher than the 57.59 achieved by transcribing first and then retrieving, and code retrieval on MTEB Code improves from 68.76 in EmbeddingGemma 1 to 78.68. There are currently no further developments.

View story:Google Releases EmbeddingGemma 2 Paper

Kimi integrates with Gmail, Google Drive, and Google Calendar

English text is being prepared.

View story:Kimi integrates with Gmail, Google Drive, and Google Calendar