AI Models Release Notes
Release notes for leading AI models, APIs and AI platforms
Products (16)
Latest AI Models Updates
- Oct 17, 2026
- Date parsed from source:Oct 17, 2026
- First seen by Releasebot:Oct 18, 2026
DeepSeek-V4.1-Flash-1017 Update
Deepseek upgrades V4.1 Flash to V4.1-Flash-1017 across its web, app, and API models, improving benchmark performance while reducing task runtime and output length for faster, lower-cost use. API access and pricing remain unchanged.
DeepSeek's web and API search capabilities are powered by our in-house DeepSeek search engine, which reliably handles hundreds of millions of search requests from around the world every day. DeepSeek Search is an AI-native general-purpose search engine that indexes hundreds of billions of web pages, supports multilingual search, and returns highly relevant, authoritative, and timely results in real time. It filters SEO spam, GEO manipulation, and low-quality AIGC through strict detection mechanisms, ensuring authoritative and trustworthy information for users and LLMs. For usage details, see the Search & Fetch guide.
DeepSeek V4.1 Flash has been upgraded to DeepSeek-V4.1-Flash-1017. The official web and APP models have also been updated to the new version. The API calling method and pricing remain unchanged — simply use deepseek-flash to access the latest model.
Performance–cost Pareto improvement
Compared with the previous DeepSeek V4.1, the new model delivers substantial improvements across benchmarks:
- HLE (w/o tools): 42.6 (44.0*)
- Terminal-Bench 4.0: 42.3
- DeepSWE v1.1: 77.0
- ProgramBench: 29.1
- NL2Repo-Bench: 68.2
- Automation-Bench v1.0.6: 60.5
- Agents' Last Exam: 35.4
- Apex Bench v1.1: 72.1
- Chartography (w/tools): 83.0
- BabyVision (w/tools): 92.5
- ZeroBench-main (w/tools): 51.0
- Tested only on the pure-text subset of the HLE benchmark set. Terminal-Bench 4.0 was tested with the Claude Code framework, and DeepSWE with the Mini-SWE-V2 framework. ZeroBench-main was evaluated with Pass@5.
While improving performance, we also reduced the model's task runtime and output length, enabling users to complete tasks at lower cost and higher speed and achieving a new performance–cost Pareto optimum across multiple harnesses. We specifically optimized for DeepSeek Harness, Tencent CodeBuddy, WorkBuddy, and other platforms. Please upgrade to DeepSeek Harness 0.2.1-rc.1 or later for the best experience with the new model.
Original source - Oct 16, 2026
- Date parsed from source:Oct 16, 2026
- First seen by Releasebot:Oct 17, 2026
October 16, 2026
ChatGPT launches Slides for creating, editing, presenting, and collaborating on decks, with export options across web and desktop. It also updates GPT-6 Pro with faster responses, improved tool use, and parallel agents, rolling out to eligible users.
Create and collaborate on slides in ChatGPT
ChatGPT Slides lets you create presentations and keep working on them with your team in Space. Start from a conversation, files, or an outline, or share an example deck to guide the style and layout.
- Create and refine. Edit text, images, shapes, and charts directly, or ask ChatGPT to make changes. Generate custom images or add your own.
- Work together. Invite teammates to edit the same deck in real time and leave shared comments.
- Present or export. Present directly from ChatGPT. On the web, export to PowerPoint, PDF, or Google Slides. In the desktop app, open your deck in PowerPoint or Keynote when those apps are installed.
Slides is rolling out to eligible Pro, Business, and Enterprise users with Space access on the web and in the desktop app. To get started, ask ChatGPT in Work to create a presentation in Space.
Updates to GPT-6 Pro model in ChatGPT
GPT-6 Pro in the chat tab now offers:
- Faster responses
- Improved use of plugins and tools to carry out tasks.
- Multiple agents that can work on different parts of a task in parallel. You can follow their progress in the conversation and open completed tasks to see what each agent did.
This update is rolling out to eligible Pro users in ChatGPT. Availability in managed workspaces depends on workspace settings and rollout eligibility.
Original source All of your release notes in one feed
Join Releasebot and get updates from Deepseek and hundreds of other software products.
- Oct 15, 2026
- Date parsed from source:Oct 15, 2026
- First seen by Releasebot:Oct 16, 2026
Teaching students to think and learn through questions
OpenAI brings its AI for Educators course to Brazilian Portuguese and highlights Study Mode in ChatGPT, which guides learners through topics with questions, hints and explanations while supporting teachers’ classroom work.
On Teachers’ Day, we celebrate those who help us learn. A good question can change the course of a lesson. It can bring an unspoken question to light, open up another way to solve a problem, or help someone recognize what they already know. Teachers do this work every day: they know their students, choose how to introduce an idea, and guide them from their first attempt to understanding.
Set in a university classroom, the campaign takes a humorous look at the changing relationship between teachers and students and celebrates the role of educators in a world with AI.
AI for Educators now available in Portuguese
Today, we are launching the AI for Educators course in Brazilian Portuguese on OpenAI Academy, OpenAI’s free learning platform. Designed for educators in primary, secondary, and higher education, the course shows how to use AI to plan lessons, create activities and assessments, and prepare communications while preserving educators’ pedagogical judgment.
Learning through active participation, step by step
Study Mode helps you explore a subject step by step, with questions, hints, explanations, and checks for understanding. Developed in collaboration with teachers, scientists, and experts in pedagogy, the feature aims to encourage active participation and reflection on one’s own reasoning.
To turn on Study Mode, select Study from the tools menu in ChatGPT.
A student reviewing biology, for example, might ask: “I’m studying enzymes. Ask me one question at a time, wait for my answer, and help me identify what I need to review.”
Teachers can also try Study Mode with subjects they know well, looking at whether the questions encourage reasoning and whether the explanations are appropriate for their class.
In a science lesson on evaporation, for example, a teacher might ask students to investigate why clothes dry on a clothesline. After exploring the topic with Study Mode, each student writes their own explanation and a question to discuss in class, without referring back to the conversation. The teacher can use these responses to identify what needs to be revisited with the class.
This Teachers’ Day, we invite you to experiment: explore the AI for Educators course, try Study Mode, and discover how these tools can be useful in different teaching contexts. Technology can open up new ways to learn, but it is teachers who know their students, ask the most important questions, and help turn curiosity into knowledge.
Original source - Oct 15, 2026
- Date parsed from source:Oct 15, 2026
- First seen by Releasebot:Oct 16, 2026
Code Review is now generally available
Claude brings Code Review out of research preview for Team and Enterprise, with reviews that finish about three times faster, clearer bug findings, and Quick, Standard, or Deep settings for each repository. Reviews run automatically on new pull requests once enabled.
Catch more bugs before they reach production.
Code Review is out of research preview and available to Claude Team and Enterprise organizations.
Eligible Claude Enterprise organizations can claim at least $10,000 in Code Review credits by January 15, 2027. Credits expire 30 days after you claim them. Terms apply.
What's new in Code Review
Coding agents keep increasing the pace at which teams write code, and the review bottleneck hasn't eased. For general availability, we made Code Review faster, better at finding the bugs that matter, and easier to tune to each of your repositories.
Since we released the research preview, Code Review has gotten sharper, catching more of the bugs that would otherwise reach production. We also run Code Review on pull requests at Anthropic, which shows us where reviews slow our own engineers down. Teams using Code Review during the preview told us the rest: it was catching real issues before a human reviewer looked at the PR, but they wanted reviews that were quieter, sized to each change, and faster. Here’s what we shipped:
- Speed: Most reviews now finish about 3x faster (~20 minutes down to 6-10 minutes) compared to during preview.
- Quality: Find more of the bugs that matter, with clear, concise findings that state the problem, what triggers it, and the fix in a few sentences.
- Flexibility: Admins choose Quick, Standard (default), or Deep for each repository, so routine codebases get a quick pass and the most critical ones get a deep review. Every level delivers high-quality reviews.
Get started
Code Review is now generally available for Team and Enterprise plans.
- For admins: Enable Code Review in your Claude Code settings, install the GitHub App, and select repositories you’d like to run reviews on.
- For developers: Once enabled, reviews run automatically on new PRs. No configuration needed.
Explore the docs for more information.
Original source - Oct 15, 2026
- Date parsed from source:Oct 15, 2026
- First seen by Releasebot:Oct 16, 2026
October 15, 2026
ChatGPT adds GPT-Live-1.1 with more natural accents, new voice options and clearer search cues, while removing the fixed cap on Sites creation. Applications also open for eligible U.S. university labs seeking Rosalind Academic Plan access.
GPT-Live-1.1 and new voices in ChatGPT
We’re rolling out GPT-Live-1.1 and GPT-Live-1.1 Quip, a mini version of the model, in ChatGPT Voice, with more natural accents across languages. Live also uses a wider range of short phrases to let you know when it’s searching.
New Live voices include options for French, German, Hindi, Japanese, Korean, and Mexican Spanish, plus Bubbie in English. The voices you see may vary by region and account.
To choose a voice, open Settings > Voice, then select Voice.
For more information, see: ChatGPT Voice.
Applications for the Rosalind Academic Plan
The Rosalind Academic Plan helps eligible academic labs use ChatGPT for life sciences research. Approved labs can receive up to 30 ChatGPT Business Standard seats at no cost for up to 12 months, with GPT-Rosalind access subject to trusted-access approval.
Applications are now open to labs at U.S. universities. If you’re a principal investigator or lab lead, apply on behalf of your lab. Applying doesn’t give your lab immediate access. After approval, follow the instructions to set up your workspace and enable access for approved members. Labs covered by an OpenAI Enterprise agreement should use their university’s access process. For eligibility and setup details, see: Rosalind Academic Plan.
Site limits (Plus and Pro)
There is no longer a fixed limit on the number of Sites you can create. Usage and storage limits continue to apply.
For usage and limits, see: Creating and using ChatGPT Sites.
Original source - Oct 15, 2026
- Date parsed from source:Oct 15, 2026
- First seen by Releasebot:Oct 16, 2026
Introducing Timestral
Mistral releases Timestral, its first family of time-series foundation models, with four sizes for probabilistic, multivariate forecasting and support for past-only and known-future covariates. The models lead public benchmarks, run on standard CPUs, and are available to try through a published GitHub repository.
Today, we’re releasing Timestral, our first family of time-series foundation models. Our largest model, at just 209M parameters, ranks first on all three of the field's main public benchmarks, GIFT-Eval, TIME, and fev-bench, outperforming the previous state of the art with only two-thirds of the model parameters.
Timestral delivers probabilistic, multivariate forecasts and supports past-only and known-future covariates. Our family of models sits on the Pareto front of forecasting accuracy versus model size, delivering leading performance at each size. Results have been submitted to the GIFT-Eval and TIME leaderboards.
Time Series Foundation Models (TSFM)
Time series is one of the most common data types in enterprise, from sensor telemetry to observability metrics. Forecasts built on it drive decisions in capacity planning, energy, supply chains, risk and maintenance.
Industrial plants forecast process sensors against the production schedule, catch equipment drift before it fails, and plan capacity and spare parts across thousands of assets. Finance teams forecast flows, volumes, and volatility to inform positions, hedging, and treasury across thousands of accounts.
Classical models such as ARIMA and linear regression remain well-established tools for forecasting time series. These models are usually fitted to a single series and learn only from that series' history.
Deep learning models such as PatchTST add the capacity to learn nonlinear patterns across many related series, but they are still typically trained from scratch on each dataset, so what they learn doesn't carry over to new domains.
Time series foundation models bring the pre-training approach to time series. Trained on a large and diverse corpus, our model learns patterns that transfer across series, forecasting zero-shot on data it has never seen, even in domains where data is scarce.
Benchmark Results
We evaluated Timestral on the three main public forecasting benchmarks: GIFT-Eval, TIME and fev-bench. Among pre-trained, zero-shot models, it ranks first across all five metrics. These results come from a single checkpoint, without ensembling.
GIFT-Eval, top ten of 83 comparable pretrained and zero-shot entries. Lower is better.
TIME, top ten of the 35 leaderboard entries. Lower is better on both metrics.
fev-bench, top ten of the 37 leaderboard entries, by SQL skill score. Higher is better.
Family of models optimised for every specification
Timestral has been optimised as a family of four models: 5M, 28M, 123M and 209M parameters. The larger models improve forecast performance, while the smaller ones are faster and cheaper. All four share the same architecture and training recipe. Every size sits on the Pareto front of accuracy versus model size: for each model size, no public model forecasts better.
Score vs model size on GIFT-Eval, TIME and fev-bench (shown as 1 − skill score). Lower is better. Our 28M model beats Chronos-2, about four times larger, on GIFT-Eval and TIME. Our 209M model beats TimesFM-3 and Falcon-2.0 with fewer parameters.
Timestral does not need a GPU at inference. On a standard server CPU with 8 threads, forecasting one day ahead from three weeks of 15-minute history takes 10 ms with the 5M model and 140 ms with the 209M model. Smaller models trade some accuracy for speed: the 28M model is five times faster than the 209M model, and the 123M model comes within 0.2% of the 209M score at three quarters of its latency.
An industrial example
Consider an industrial plant with thousands of sensors across its equipment. Timestral can improve the forecasts on that data in several ways.
Multivariate forecasting
A pump's vibration, temperature and pressure don't move independently. A rise in one often explains or causes a change in another. Timestral models related series jointly, so it can use those dependencies instead of forecasting each sensor in isolation.
Probabilistic forecasting
A single predicted value may not show the full picture. Timestral predicts a full set of quantiles, so teams can see how wide the uncertainty is, set alert thresholds that reflect it, and plan maintenance around risk rather than a single number.
Categorical covariates
Some covariates take values from a set of categories rather than continuous numbers, such as the day of the week, or a machine's state (running, idle, under maintenance). Encoding these as numbers would impose an order or scale they don't have. Timestral takes them as categorical inputs directly to improve forecasts.
Known-future covariates
Some drivers of sensor behaviour are known in advance, such as production schedules, planned shutdowns, weather forecasts. Timestral can condition on them to improve forecast quality, so a scheduled load increase is expected rather than flagged as a surprise.
From Model to System
Timestral is a numerical model that can run inside existing monitoring pipelines, or be paired with our language models in multi-modal agentic workflows.
In the plant, that could look like this:
- Timestral running continuously on the plant's sensor streams, detecting anomalies as they occur.
- When it flags an anomaly on a pump, an agent embeds the signals and compares against embeddings of past sensor data to find similar historical failures.
- The agent then starts sub-agents to run a root cause analysis, drawing on context from that plant, such as maintenance records, shift logs and operator notes, equipment manuals, past work orders, and readings from related equipment.
- The agent drafts a suggested diagnosis based on the findings, with supporting evidence and possible next steps.
The plant manager gets a diagnosis and a plan backed by the sensor data, rather than just an alert.
Key technical contributions
Several ideas have been combined to produce these results. Each will be described in detail in the technical report.
- Decoupled Horizon Forecasting. Causal decoders trained with contiguous patch masking, as in Toto 2.0 and TimesFM-3, mask out the values to be forecast, so they look like missing values and are also lost as context for later forecasts. Our approach, DHF, separates the two: the full history stays in the sequence, and learned forecast tokens read the history before them. Every observed value thus serves as both target and context, true gaps stay flagged as missing, and training horizon lengths and missing-data augmentation can be tuned independently of each other.
- Covariate-aware attention. Each input series is assigned a role: target variate, past-only covariate or known-future covariate. Known-future covariates are visible across the whole window while targets stay causal, and information flows only from covariates to targets. Timestral can therefore use a holiday calendar or a weather forecast without leaking a target's own future into its forecast.
- Synthetic-first data. The majority of the training data was synthetic. We first generated univariate processes with coherent temporal dynamics, including trends, seasonality and diverse noise patterns. We then used a multivariate synthesizer to construct multivariate series by introducing nonlinear dependencies between these processes.
- Two-stage training. Timestral is first pre-trained at scale on purely synthetic data with context size 8192, then post-trained on a mix of real and synthetic data with a 4x longer context. The objective is a pinball loss over 21 quantiles, a discrete approximation of CRPS.
A model that can be adapted
Public benchmarks measure general performance, but production data can be significantly different in behavior and structure. It may also shift over time as markets change regime, equipment degrades and demand patterns move. The lasting value lies in being able to adapt a model to in-domain data and keep retraining it as the distribution moves.
For organisations that want to go further, Timestral is available through Mistral Forge, the same training pipeline we use for our language models. It supports:
- Zero-shot inference out of the box.
- Architectural extensions, such as domain-specific covariates.
- Fine-tuning on proprietary data, specialising to a domain while preserving general performance.
- Control over the training data, including adjusting the training mix, replaying pre-training data, and generating synthetic data samples to reflect the target domain. Full control over the data ensures that nothing we backtest was seen during pre-training.
The full pipeline can be deployed on our clients’ hardware, including air-gapped environments.
What’s next
We have published a GitHub repository so you can run Timestral on your own data. A technical report with the full results will follow. For more specialised forecasting problems, or to collaborate on research, please reach out to us.
Timestral
- Type: Time-series foundation model (TSFM)
- Parameters: 5M, 28M, 123M and 209M
- Context length: 32,768 time steps
- Output: Probabilistic forecasts, 21 quantiles
- Inputs: Multivariate targets, past-only and known-future covariates
- Deployment: On-prem, VPC or air-gapped
- Licence: Modified MIT
- Oct 15, 2026
- Date parsed from source:Oct 15, 2026
- First seen by Releasebot:Oct 16, 2026
Bringing GPT‑Rosalind to academic research labs
OpenAI launches the Rosalind Academic Plan, giving eligible U.S. university labs up to 30 free seats for a year, access to frontier models, and the option to request trusted access to GPT‑Rosalind for life sciences research.
Introducing the Rosalind Academic Plan, offering eligible labs sponsored access to GPT‑Rosalind and tools for life sciences discovery.
We’re introducing the Rosalind Academic Plan to help more academic scientists access GPT‑Rosalind, our dedicated life sciences model, which combines frontier reasoning with Rosalind workbench that enables specialized tool orchestration across medicinal chemistry, genomics, wet-lab assistance, and other scientific applications. Eligible labs can receive up to 30 Business Standard seats at no cost for up to 12 months, with access to GPT‑Rosalind subject to trusted-access approval.
Academic research creates knowledge that entire fields build on. Work that begins with a fundamental question about biology can eventually change how we diagnose disease, develop medicines, or grow food. But those advances depend on scientists having the resources to explore ideas, test them, and share what they learn.
Accelerating scientific discovery is central to OpenAI’s mission of ensuring that artificial general intelligence benefits all of humanity. With this plan, we’re investing in the scientists doing that foundational work—helping more labs combine their expertise and judgment with AI tools built for life sciences research.
Supporting the research behind future breakthroughs
Many of the advances we rely on today grew out of years of basic research.
The Human Genome Project helped launch modern genomics, transforming medicine, diagnostics, and biotechnology.
Curiosity-driven work on bacterial immune systems led to CRISPR-Cas9, which revolutionized genome editing and is now advancing research in cancer, inherited disease, agriculture, and biotechnology. Decades of university research on mRNA also laid the groundwork for COVID-19 vaccines and a new generation of vaccines and therapies.
Academic research has always laid the groundwork for advances in medicine and biotechnology. By helping more scientists combine their scientific expertise and judgment with the advanced specialized reasoning of GPT‑Rosalind, we hope to support discoveries whose benefits reach far beyond the lab.
How academic labs are using GPT‑Rosalind
OpenAI is committed to enabling academic researchers to pursue their scientific ambitions. Distinguished academic labs are already exploring how GPT‑Rosalind can support work ranging from protein design to understanding neurodegeneration. These examples show where researchers are applying the model within their scientific workflows.
Zhao Lab
UCSF
Quantitative Drug Development
Gathering drug-property evidence to support modeling how medicines move through the body and produce effects.
“GPT‑Rosalind helped me gather drug physicochemical parameters faster. PubChem and NCBI tools helped me work across ten drugs while refining the distinction between predicted and experimentally measured values”
Zhao Lab, UCSF
What the plan includes
With the new Rosalind Academic Plan, academic researchers can bring GPT‑Rosalind and Rosalind Workbench to their lab with greater flexibility and harness frontier model intelligence for scientific research.
Eligible labs receive:
- Up to 30 Business Standard seats at $0 per user per month for up to 12 months.
- Access to all commercially available frontier OpenAI models, subject to plan usage limits.
- Access to GPT‑Rosalind following trusted-access approval.
Research data shared through the Rosalind Academic Plan is not used to train our models by default.
Build for life sciences research
GPT‑Rosalind is a specialized life sciences model designed to support deeper reasoning on complex biological questions. Through trusted access, approved researchers can use it to investigate sensitive scientific questions within safeguards designed for legitimate research.
Areas of research include:
Drug and protein design
Designing candidate drugs, therapeutic proteins, peptides, and enzymes, and optimizing their molecular interactions, selectivity, stability, and suitability for development.
Apply on behalf of your lab
The Rosalind Academic Plan is currently available to academic labs at U.S. universities. PIs and lab leads at universities without an OpenAI Enterprise agreement can apply on behalf of their labs through our trusted access process. Labs at universities with an existing agreement must apply through their university’s Enterprise access process.
Once approved, PIs and lab leads can activate free access to GPT‑Rosalind for up to 30 named members, alongside the other OpenAI frontier models included in the plan. API access is not included at this time. Verified PIs and lab leads manage member permissions and are responsible for their lab’s approved use.
Get started with GPT‑Rosalind
Choose the application path for your lab or organization.
Academic labs
Rosalind Academic Plan
For PIs and lab leads at eligible organizations without an existing Enterprise account.
Apply on behalf of your lab.
Existing Enterprise accounts
GPT-Rosalind access
For academic users at eligible institutions with an existing OpenAI Enterprise agreement.
Request access through your existing Enterprise account.
GPT‑Rosalind access is subject to approval.
Learn more about eligibility, application requirements, and getting started in our help center guide.
Responsible access for scientific research
Before activating access, we:
- Verify the institution’s legitimacy and the PI or lab lead’s identity
- Review the lead’s professional background, institutional appointment, and authority to oversee lab access
- Assess the lab’s institutional affiliation, research activities, and intended use
- Establish clear responsibilities for the lab lead and a designated institutional safety contact
Safeguards continue after activation through monitoring and enforcement. For registered high-sensitivity use cases, our safety checks account for the approved research context. When questions arise, we may follow up with researchers to clarify the work and confirm authorization. This approach is designed to detect potential misuse while accounting for legitimate research context and reducing unnecessary interruptions to authorized work.
Specialized intelligence for complex scientific questions
Designing better antiviral therapeutics
GPT-RosalindPrompt
I’m planning a second-generation lipopeptide campaign for a Nipah F HRC-mimetic lead that is potent on NiV-Malaysia pseudotypes but drops roughly 40-fold on NiV-Bangladesh and Hendra, gets clipped in airway surface liquid, and is weaker after live-virus attachment. Could you propose a 20-variant sequence/presentation matrix, the NiV-M/NiV-B/HeV HRN-HRC interfaces we should model, the paired pseudovirus and authentic-virus assays to run, and the artifact checks we’d need to separate true loss of F engagement from entry-route, aggregation, or serum-binding effects?
Response
Designing better antiviral therapeuticsGoal
Compare second-generation antiviral peptide candidates and explain the lead’s uneven performance. The scientific question is whether the reported potency loss reflects weaker fusion-target engagement, different presentation at the membrane, or loss of intact and available peptide in biological fluids.
Overview of readouts
The proposed readouts include IC50/IC90 and neutralization endpoints, binding and fusion signals, and peptide integrity and retained activity in airway fluids. The proposed comparison spans surrogate and authentic-virus systems and considers aggregation, serum binding, and cell-viability signals separately. This separates antiviral potency from a direct affinity measurement, and loss of engagement from loss of intact or available compound. It also distinguishes the intended biological effect from interference with the readout itself.
Assessment
It provides a candidate comparison matrix, a modeling plan, and paired study plans, with controls and interpretation guidance that address the lead’s uneven performance.
Mainline AstraPrompt
I’m planning a second-generation lipopeptide campaign for a Nipah F HRC-mimetic lead that is potent on NiV-Malaysia pseudotypes but drops roughly 40-fold on NiV-Bangladesh and Hendra, gets clipped in airway surface liquid, and is weaker after live-virus attachment. Could you propose a 20-variant sequence/presentation matrix, the NiV-M/NiV-B/HeV HRN-HRC interfaces we should model, the paired pseudovirus and authentic-virus assays to run, and the artifact checks we’d need to separate true loss of F engagement from entry-route, aggregation, or serum-binding effects?
Response
I need to prioritize safety over the requested optimization matrix. A high-level comparative framework may be more appropriate than actionable variant-design guidance.
I can provide a non-operational study framework, but not sequence-level designs or implementation-level virus protocols. Those details would provide actionable assistance involving high-consequence pathogens.
Authentic-virus studies require an appropriately authorized high-containment program.
Assessment
The response withholds the actionable parts of the request, including specific candidate designs and implementation-level experimental plans. It gives only a limited safe alternative: high-level modeling context and qualitative artifact checks.
Rosalind Goal and Overview sections are edited summaries from the original Rosalind final answers. Astra Thinking sections are edited excerpts from supported Responses API reasoning summaries returned with the new Astra reruns. Astra Final Answer sections are edited excerpts from the corresponding final answers. Assessments address textual fulfillment of requested deliverables and do not verify scientific correctness, feasibility, or whether a method works.
Get started
We’re starting with academic labs as part of a broader effort to put useful AI tools in the hands of more scientists. By lowering the cost of access, we hope to give more researchers room to pursue ideas whose benefits could reach far beyond their own labs—from a clearer understanding of disease to better treatments and more resilient food systems.
Those advances will come from scientists asking important questions and doing the work to answer them. We want to help more of them do that work.
If you’re affiliated with an academic lab, learn more about the Rosalind Academic Plan and check whether you’re eligible.
Original source - Oct 15, 2026
- Date parsed from source:Oct 15, 2026
- First seen by Releasebot:Oct 16, 2026
GPT-Live-1.1 is rolling out with new voices and better accents
OpenAI releases GPT-Live-1.1 with 13 new voices, stronger multilingual performance, and more natural delegation for ChatGPT, Codex, and dots.
New voices. Better accents. And less “checking…”
GPT-Live-1.1 is an update for ChatGPT, Codex, and dots, introducing 13 new voices and considerably better performance in French, German, Hindi, Mexican Spanish, Korean, Japanese, and many other non-English languages.
Delegation is now more natural, and steerability has also improved.
Continuous improvements to voice models and the overall product experience will become the norm going forward.
Rolling out today to all users. Coming soon to the API.
Original source - Oct 15, 2026
- Date parsed from source:Oct 15, 2026
- First seen by Releasebot:Oct 15, 2026
Accelerating the search for antimicrobial peptides with Google Antigravity
Gemini highlights Google Antigravity’s new Biomolecular Structure Viewer, a shared 3D workspace for researchers and AI agents to inspect protein structures and annotations while exploring antimicrobial peptide candidates.
New antibiotics are urgently needed to help with rising levels of antimicrobial resistance, but searching for new therapeutics is often a slow, labor-intensive process. With Google Antigravity, our agentic development platform, this process can now be made easier.
Google Antigravity provides an agentic science workbench that empowers biologists to discover, structurally evaluate, and prioritize promising targets quickly, compressing what used to be a fragmented search over multiple databases into a single session. It also helps researchers scan through millions of previously uncharacterized proteins in the UniProt database using ProtNLM, a model that predicts biological function from amino acid sequences.
This technical walkthrough shows how biologists used Google Antigravity to uncover a novel antimicrobial peptide candidate which has predicted activity against the pathogens involved in respiratory tract infections. The lead identified here is now undergoing wet-lab validation.
Finding new antimicrobial peptides
Rising antimicrobial resistance is one of the most pressing global health crises of our time and could result in an estimated 10 million annual deaths globally by 2050. To prevent this, we urgently need to discover novel therapeutics. Antimicrobial peptides (AMPs), particularly bacteriocins, offer a promising alternative to traditional antibiotics because they have naturally evolved to target fundamental vulnerabilities in competing bacteria, from cell envelopes to essential intracellular machinery, making it difficult for pathogens to develop resistance.
The exponential growth of genomic and metagenomic sequencing has populated global databases with billions of natural amino acid sequences. This pool of natural data has already produced globally used bacteriocins like Nisin, alongside recent genome-mined breakthroughs such as Darobactin, Dynobactin, and Lariocidin.
Many, however, remain locked within the unannotated "dark matter" of protein databases, with their function uncharacterized because they aren’t similar to known proteins. For example, the world's leading public resource of protein sequence and functional information, UniProt, holds millions of sequences with unknown functionality, transferred directly from genome sequencing projects.
To bridge this gap, Google developed ProtNLM (Protein Natural Language Model), a deep learning model trained to directly read the underlying evolutionary “grammar” of amino acid sequences. This allows the model to predict biological function directly from raw protein sequences, enriching millions of previously unnamed UniProt entries with high-confidence functional labels.
Overview of the ProtNLM pipeline translating raw amino acid sequences into natural-language biological annotations for unreviewed UniProtKB/TrEMBL entries. The functional annotation scope of the system covers protein names, function comments, subcellular locations, and GO terms, in addition to keywords.
Collaborative intelligence
Scientific discovery is rarely a single step process; finding viable candidate proteins requires human intuition and iteration to incorporate new empirical evidence. Working in Antigravity, the researcher and AI agent begin by setting a focused biological goal: find and computationally validate novel AMPs in Streptococcus pseudopneumoniae. During an interactive, multi-turn session the researcher steers the scientific strategy, deciding which questions to ask and what evidence to collect, while the agent handles the heavy lifting of cross-database orchestration, structural analysis, and literature synthesis.
Navigating a vast search space: The agent queries the database and filters millions of entries down to a handful of candidates worthy of attention. Here, the agent finds a bacteriocin (Uniprot ID: A0A0U0D591) with a length of 97 amino acids from the Lactococcin 972 family.
The candidate we uncover in this walkthrough represents a biologically plausible lead backed by structural and genomic evidence but it still requires wet lab validation. What makes this meaningful is how dramatically it lowers the barrier to everyday biological workflows, turning days of fragmented database triage into a single collaborative session. While this session focused on a narrow amino acid window in Streptococcus pseudopneumoniae, researchers can take the conversation in any direction, such as layering in new filtering criteria, exploring different sequence lengths, or targeting other priority pathogens. By broadening the search from human commensals to environmental microbes and beyond, scientists can guide the agent across UniProt to discover novel AMP candidates at scale.
A shared 3D canvas for interactive collaboration
A protein's 3D structure can help us understand its biological function. Examining a candidate’s predicted fold in space is key. To provide this end-to-end workflow in a single workspace, we are introducing the Biomolecular Structure Viewer, a 3D visualization tool embedded directly into Antigravity.
The researcher and agent work together over a shared visual state. As the agent retrieves structures from AlphaFold Database or the scientist’s own predictions, it can dynamically drive the viewer. For instance, it can highlight key functional regions to visually explain its hypotheses. Crucially, this awareness flows both ways: actions like highlighting specific residues or pinning notes directly onto a binding pocket are synchronised back to the agent. A researcher can simply mark a region of interest in 3D space and ask, for example, "How might this β-strand affect Lipid II binding?", allowing the agent to immediately reason over the exact structural context the scientist is looking at.
The Biomolecular Structure Viewer inside Antigravity showing the AlphaFold-predicted structure of the candidate bacteriocin, giving the researcher and agent a shared 3D canvas to inspect and annotate functional regions.
Accelerating Scientific Discovery
Historically, uncovering insights like those in the workflow demonstrated above required fragmented and specialised software: writing custom scripts to query databases, manually exporting files to 3D modelling programs, managing disconnected genomic tools and working with multiple experts to design the wet lab assays. Because wet-lab validation remains resource-intensive, narrowing the search space to only the highest-confidence leads upfront is critical. Using Google Antigravity allows researchers to navigate the vast biological search space efficiently while leveraging accurate protein function predictions from a domain-specific model.
Google Antigravity executes this end-to-end scientific reasoning in one place. This allows scientists to spend more time evaluating high-quality hypotheses, running experiments in the lab and solving research challenges.
Acknowledgements
We would like to acknowledge the contributions of the following team members: Maria Gabriela Cardoso, Victoria Johnston, Lexi Walker, Gena Hong, James King, Richard Green, Filip Radlinski, Artem Sobolev, Michael V. LeVine, Ruijie K. Zhang, Zachary Wu, Luca Schulz, Amy Danson, Tristan Oliver Kwan, Ryan Krueger, Oleg Kovalevskiy, Doug Fritz, Yunhan Xu, Sebastian Nowozin
Original source - Oct 15, 2026
- Date parsed from source:Oct 15, 2026
- First seen by Releasebot:Oct 15, 2026
An intelligent new shopping cart
Gemini brings Google’s Universal Cart to U.S. shoppers, helping them save products from multiple retailers across Search and the Gemini app, track price and stock changes, find savings, and check out with Google Pay or on merchant sites.
Universal Cart launches in the U.S. today. Here are four ways it can help with holiday shopping — including organizing products from multiple stores into a single cart, tracking price drops, and more.
As we build for the future of commerce, our focus spans two fronts: unlocking agentic capabilities that act on your behalf, and designing practical tools that give shoppers confidence and control. Google’s Universal Cart is built for that second track — mirroring how people actually shop today across Google, without losing track of their items, their savings, or the merchants they trust.
The Universal Cart, which we previewed earlier this year, is an intelligent shopping cart and your new hub for shopping on Google. The cart puts you in control, helping to organize your shopping, keep track of items you love from across retailers, stay on top of promos, and enjoy effortless checkout powered by the Universal Commerce Protocol (UCP). Today, Universal Cart is rolling out in the U.S. across Google Search and in the Gemini app.
The cart for all your shopping
Here are four things the new Universal Cart lets you do to stay organized and shop more confidently this holiday season.
Save products from multiple retailers, wherever you’re shopping on Google.
Universal Cart offers a more organized alternative to tabs and screenshots. The cart saves your progress across Google so you never have to start your shopping from scratch. Let’s say you save a handbag from Steve Madden to your cart in AI Mode in Search, and later a sweater from Quince and lipgloss from Ulta Beauty in the Gemini app. Those products will be there in your cart when you come back later so you can pick up right where you left off.
Keep tabs on pricing or inventory changes for your entire list.
The cart is built on the 60+ billion continually refreshed product listings in Google’s Shopping Graph. This allows it to proactively work for you in the background, sending notifications about price drops or when an item comes back in stock.
Identify hidden savings or points from your cart.
The cart’s intelligent reasoning can help anticipate your needs. Say you have a couple different credit cards, but don’t know each one’s specific benefits. The Universal Cart is built on Google Wallet so it understands your card perks and can intelligently help you choose between payment methods. It’ll also show you relevant promo codes right from the cart.
Make seamless, trusted purchases when you’re ready to buy.
In online shopping, trust is everything. You can securely check out with Google Pay in just a few taps with many of your favorite stores right in the cart, or you can transfer your items directly to the merchant’s cart to finish buying on their site. Try these simplified checkout experiences with merchants like Nike, Sephora, Target, Ulta Beauty, Walmart, Wayfair, and many Shopify merchants. Either way, you will receive direct confirmations and customer support from the merchant, and they will remain the seller of record.
As the holiday season kicks into high gear, the Universal Cart is just one way we’re giving shoppers the convenience and smarts they need to make the best shopping choices.
Original source