Capturing token IDs during agentic interactions for better reinforcement learning
Reinforcement learning (RL) is one of the techniques we use to make language models better at sustained, multistep tasks like writing code, navigating a website, or carrying out a research workflow. The model doesn't act alone in those settings; it's wrapped in a piece of software we call a harness, which lets it call tools, observe the results of using them, and decide what to do next. To improve such a model with RL, we let it attempt many tasks inside the harness, score how well each attempt went, and use the score to nudge the model's parameters toward the choices that worked. The hard part turns out to be the bookkeeping. To turn a scored attempt into a parameter update, the trainer needs an exact record of what the model produced — not a summary, and not a transcript that looks as though it captures a complete exchange but drops vital information. Internally, models see text as a sequence of numbered units called tokens; an English sentence might have 10 or 20 tokens, each assigned an integer ID by a piece of software called a tokenizer. Two strings that look identical in a transcript can map to different token IDs after a small change of formatting, and that gap, however small, is enough to make the trainer optimize against a slightly different past than the one the model actually experienced. Today we're releasing Turnstile, a small proxy written in the Rust programming language that sits between any agent harness and the backend system that runs the model. Turnstile records the exact token-level history of every request as it happens, at the only point where that history is unambiguously correct: the moment of generation. It then exports a generic, framework-neutral trajectory that can feed into whichever RL training stack you already use. We use Turnstile to drive real RL training runs. In the validations we report here, two different agents — a text-only coding agent and a multimodal computer-use agent — improved steadily over the course of their RL runs. In both cases the agent harness was left unchanged, and the data Turnstile records flowed directly into the training stack and produced the expected learning signal end to end. Tokens, rollouts, and why agent transcripts can lie Three pieces of vocabulary do most of the work in the rest of this post, so it's worth grounding them up front. A tokenizer is a deterministic function that turns text into a list of integer token IDs and token IDs back into text. Each model is paired with a specific tokenizer; you can't substitute one for another. The tokenizer is unforgiving: a stray space, a different way of writing a tool call as JSON, or a slightly different chat template (the format string a serving system uses to wrap roles and messages into a single text input the model can read) can change the token IDs even when the rendered text looks the same to a human. We'll call this kind of mismatch retokenization drift when it comes from rerunning a tokenizer over text we’ve already seen and chat template drift when it comes from the surrounding format changing under us. A rollout is one recorded attempt at a task: the prompt, every tool call, tool feedback, the model’s responses, and the final outcome. The version of the model that produced the rollout is called the behavior policy. The mathematics of policy-gradient RL works cleanly only when the trainer optimizes the model's behavior against the context the behavior policy actually saw. If we rerender the prompt and end up with a slightly different token sequence, we're now training the model against a context unfamiliar to the behavior policy. The training signal degrades, sometimes invisibly, since the model still appears to be learning. That is why agent harnesses make the problem worse rather than better. A harness is not a static prompt; during a single rollout it may compact older messages to save context, retry a malformed tool call, branch into subagents, merge their results back, or summarize history. All of that is normal, useful agent behavior. But each rewrite is another chance for the next request's token sequence to drift away from what the model actually generated last turn. The transcript a harness produces is a faithful record of the conversation; it is not, in general, a faithful record of the tokens, and it is the tokens the trainer needs. Capturing tokens at the proxy boundary Turnstile's central design choice is to stop trying to reconstruct token-level state from text after the rollout is over. We capture it at the moment of generation, where it is already correct, and we do that without changing the harness. The proxy speaks the same HTTP API every modern agent harness already speaks: the OpenAI Chat Completions API, which has become the de facto standard for "send a list of messages, get back a response." The harness creates a rollout group with Turnstile, points its Chat Completions client at Turnstile's address instead of the real backend, and runs unchanged. Behind the scenes, every request flows through Turnstile to the inference backend (today SGLang, with vLLM planned). Turnstile records the exact token IDs the model sampled, the per-token log probabilities (the model's own confidence values for each token, expressed as logarithms; the trainer needs these to compute its update), and a loss mask that marks which tokens were generated by the model and should contribute to training versus which came from the user, tools, or the system prompt and should not. When the rollout is finished, the harness asks Turnstile for the recorded trajectories. Each trajectory is a TrainingSequence object containing the token IDs, log probabilities, the loss masks for the full sequence, and a record of which version of the model's weights was active during which spans of the sequence. (The trainer needs these weight version boundaries to know if the model parameters changed mid-rollout because of an asynchronous update.) Turning that into the specific batch shape your trainer wants is straightforward adapter work: attach the reward, expand the mask, and hand it off. Existing harnesses can stay black boxes There are a number of agent harnesses, such as OpenHands, Codex, and Terminus, that are already useful but were not designed as training runtimes. Without Turnstile, using one of these to drive an RL training run requires it to record token IDs, log probabilities, masks, and routing traces; in practice, that work often falls to a separate harness-shaped component built inside the training system. Either way, a harness ends up acting as a token-level RL data pipeline, and that is the wrong abstraction. The harness knows the information it intended to send to the model, but it does not, in general, know the exact token sequence, cache state, routing trace, or processed multimodal inputs the model actually used. Those live in the backend. With Turnstile, the production harness doesn't have to log training data. It points its Chat Completions client at Turnstile instead of the inference backend and otherwise runs unchanged. Turnstile records the model-facing rollout state: it does not need to understand the harness's private control flow or the semantic reason a context changed. If the next request is a faithful token-level extension of an earlier one, Turnstile merges it into the same trajectory. If the harness compressed memory, rewrote history, merged a subagent result, or otherwise changed the prefix in a way that cannot be proven equivalent, Turnstile starts a new sequence and keeps the trainable suffix honest. All this applies in the strict black-box case: a proprietary harness whose internals are closed to the training system can still drive an RL training run, with no source-level integration at all. Below are two examples of using Turnstile with open-source harnesses in a black-box fashion. Multiturn agents and prefix-aware trajectories A naïve way to store these recordings would be to treat every request as an independent training example. That’s wasteful, because each new request includes the whole conversation so far, so the same token strings would end up being duplicated over and over. It’s also subtly wrong, because it loses the relationship between turns. Instead, Turnstile stores a multiturn rollout as a single growing token path — so long as the path is faithful to what the model actually saw. When a later request to the model is just the previous request plus a few new tokens at the end (the new user message, a tool result, an LLM response), Turnstile recognizes the overlap at the token level — not by comparing rendered strings but by checking that the previously captured token IDs really do appear unchanged at the start of the new request. If they do, the two turns become one continuous trainable sequence with the loss mask correctly identifying which spans were the model's outputs. When the next request cannot be safely extended from a previous one — because the harness rewrote earlier messages, or the tokens drifted for some reason — Turnstile does not pretend otherwise. It starts a new training sequence. We call this "exploding the trajectory". It costs more training tokens than the optimistic alternative, but it ensures that every token string used for RL training is one the behavior policy actually saw. The point is not to maximize compactness; the point is never to lie to the trainer about what happened. Mixture-of-experts routing adds a hidden dimension Some modern models use a mixture-of-experts (MoE) architecture, in which only a small subset of the model's parameters — called experts — are activated for any given input token. The choice of experts is itself part of the computation, made by a small router network at every layer. The routing decision depends on the activations at each layer, and tiny differences in how the previous tokens were processed can change which experts are picked. This matters for RL because, even if two requests have the same token IDs, the tokens might get routed to different experts on different runs. Last fall, researchers at Peking University and their colleagues characterized this discrepancy and proposed recording the MoE model’s routing decisions so the trainer can replay them. We adopt the same principle. When MoE capture is enabled, Turnstile asks the inference backend for the routing trace and records it alongside the tokens. Every time it extends the token path, it checks that the routing for the shared prefix matches what was recorded last turn. If it doesn't — for example, because a key-value cache miss forced the backend to recompute the prefix and pick different experts — Turnstile splits the trajectory rather than train under the wrong routing. Multimodal rollouts When the model also takes images as input — a vision-language model — the rollout has another piece of state to keep track of. The model doesn't see the raw bytes of the uploaded image; it sees the output of an image processor, a fixed pipeline that resizes and crops the image, converts it into a tensor of pixel values, and inserts placeholder tokens into the text to mark where the image goes. The same image bytes can produce different tensors as the result of a different processor version, a different resize policy, or a different patch geometry, and the placeholder count can change with the input dimensions. If the RL trainer has to reprocess the image from scratch, the visual prefix it trains under may not be the visual prefix the behavior policy saw. Turnstile treats image processing as part of the rollout. When a request includes an image, Turnstile decodes it, hashes and stores the original bytes for audit, runs the model's configured processor, and records the processed pixel features in the same trajectory as the token IDs, in the order in which the placeholders appear. The exported sequence carries both the raw and processed visual data, so the trainer can use whichever it needs. Where this is going Turnstile is early. The current implementation has a Rust core, an SGLang backend, Python bindings for in-process training scripts, prefix-aware multiturn capture, optional MoE routing capture, and multimodal support. Near-term work is broader: a vLLM backend, more training-framework adapters, and more multimodal-model coverage. The long-term shape is unchanged from the design we started with: agent harnesses should not have to become RL data pipelines, and trainers should not have to guess what happened from rendered text. The model sampled the tokens. We record them. Additional references Turnstile is now available on Github OSWorld and its open-source agent harness OpenHands THUDM/slime Prime Intellect renderers blog Prime Intellect renderers GitHub prime-rl algorithms (extension property, multi-turn trajectory merging) prime-rl inference (router replay) Polar: Agentic RL on Any Harness at Scale NVIDIA-NeMo/ProRL-Agent-Server Stabilizing MoE Reinforcement Learning by Aligning Training and Inference Routers rLLM Agent Lightning Strands Agents SGLang provider AcknowledgementsSpecial thanks to Keagan Long, Daisy Lin, Changlong Yu, and Yifei Wang for their contributions to this work.
Event trace
Timeline
- Paul Bakaus: AMP is a pretty misunderstood project 转载
- Tested in the wild: Best survival tech! 整合
- Gmail’s Material 3 Expressive settings redesign is starting to reach Android users 整合
- 5 AI agents you can actually use on Android in 2026 (and who should use them) 整合
- I finally have Google Messages’ new pinned chats UI. Here’s how to get it on your phone now 整合
- Google Messages’ big search redesign is now rolling out to beta testers 整合
- Claude’s Duet agent is a little neurotic and surprisingly helpful 整合
- Exclusive: Google’s hidden list of 134 apps that support Gemini task automation 整合
- Motorola MA2 Adds a Phone-Switching Button to Wireless Android Auto 整合
- London hotel put admin login for booking website on whiteboard behind the front desk 整合
Time Slices
- T0 110 篇 2026-11-18 21:40 Mobile: Unit Testing Fixing Duplicate Contacts in Android People app from Google Contacts and Google+ Writing your own Karma adapter Automatic Stubbing of Network Requests to Deflakify Automation Testing Running Android UI Test Suites on Firebase Test Lab Unit Testing vs Integration Tests Increased accuracy of aapt2 "keep" rules Google AMP’s Malte Ubl wants to make the mobile web better Integration Testing for Memory Leaks Travel, Parenting, and ADHD with Heidi Waterhouse Heidi Helen is a Designer/Developer/Business Owner Selectively running Android modularized unit tests on your CI server Better Android testing at Airbnb, Part 5 Better Android testing at Airbnb, Part 6 Better Android testing at Airbnb, Part 7 Tests Under the Magnifying Lens How we enable Airbnb team members to code like a mobile engineer Airbnb’s Page Performance Score on Android Making Airbnb’s Android app more accessible Spotify and Disney have everything-app dreams 173: Mocking and Unit Tests Between hyper-focus and burnout: Developing with ADHD How Dropbox leverages testing to maintain high level of trust at scale Gustav Söderström - How Spotify Thinks A product manager’s guide to adapting Android apps across devices Now in Android #119 Getting Backstage in front of a shifting dev experience From natural language to UI tests: A deep dive into Journeys for Android Studio 用开发 Agent 理解 Agent 01:握着你的手摸索小龙虾 用开发 Agent 理解 Agent 02:数据即代码,代码即数据 Your AI Robo-Buddy Is Legally Required to Grass You Up Capturing token IDs during agentic interactions for better reinforcement learning Birgitta Boeckeler on Harness Engineering for AI Agents Eval-driven development: Lessons from evaluating GenAI at scale Flexible Authentication: Reimagining authentication for millions of users at Airbnb Bring one-handed gestures to your Wear OS app Preparing your app for broader memory limits Elevating app quality: Reducing memory usage and improving device migration Emulator control for adaptive app development Leverage Android skills and Gemma 4 in Android Studio Quail 4 Beyond the model: Engineering AI infra with scientific judgement The guest journey, updated in real time: extending Airbnb’s sequence recommender with Chronon Bring your Android game to the car screen today Land your apps on Googlebook with adaptive development Driving growth on Google Play: The next era of subscriptions Android Bench 2.0: Pushing the frontier with challenging long-horizon tasks Device Streaming and Android skills - available in Android CLI Beyond synthetic testing: Capturing and replaying real database workloads at Airbnb Introducing the Googlebook Linux Terminal Preview: A VM-Backed Environment for AI and Development Are Harnesses Worth the Cost? New ways to drive smarter discovery, deeper engagement, and business growth on Google Play Bringing our latest frontier model to Android Developers Android Developers Blog is moving Your Discord Messages Will Soon Be Discoverable on Google Search, but You Can Opt Out Why a 1B On-Device LLM Needs a Harness Google is redesigning Messages again, and heavy texters might hate it Agent Harness: Definition Android Dev Summit 2026 Gesture Controls With Only a Camera Почему мощная нейросеть может работать плохо: что такое харнесс и как он меняет возможности AI-агентов Google updates 8 Translate widgets with a clean, refined new design Henry Cavill’s Mustache Protocol Googlebook game is now ready to play Motorola’s new wireless Android Auto dongle is finally here, but it has a big price hike London hotel put admin login for booking website on whiteboard behind the front desk Motorola MA2 Adds a Phone-Switching Button to Wireless Android Auto Exclusive: Google’s hidden list of 134 apps that support Gemini task automation Claude’s Duet agent is a little neurotic and surprisingly helpful Google Messages’ big search redesign is now rolling out to beta testers I finally have Google Messages’ new pinned chats UI. Here’s how to get it on your phone now 5 AI agents you can actually use on Android in 2026 (and who should use them) Gmail’s Material 3 Expressive settings redesign is starting to reach Android users Tested in the wild: Best survival tech! Paul Bakaus: AMP is a pretty misunderstood project
- T+2h 110 篇 2026-11-18 23:40 Mobile: Unit Testing Fixing Duplicate Contacts in Android People app from Google Contacts and Google+ Writing your own Karma adapter Automatic Stubbing of Network Requests to Deflakify Automation Testing Running Android UI Test Suites on Firebase Test Lab Unit Testing vs Integration Tests Increased accuracy of aapt2 "keep" rules Google AMP’s Malte Ubl wants to make the mobile web better Integration Testing for Memory Leaks Travel, Parenting, and ADHD with Heidi Waterhouse Heidi Helen is a Designer/Developer/Business Owner Selectively running Android modularized unit tests on your CI server Better Android testing at Airbnb, Part 5 Better Android testing at Airbnb, Part 6 Better Android testing at Airbnb, Part 7 Tests Under the Magnifying Lens How we enable Airbnb team members to code like a mobile engineer Airbnb’s Page Performance Score on Android Making Airbnb’s Android app more accessible Spotify and Disney have everything-app dreams 173: Mocking and Unit Tests Between hyper-focus and burnout: Developing with ADHD How Dropbox leverages testing to maintain high level of trust at scale Gustav Söderström - How Spotify Thinks A product manager’s guide to adapting Android apps across devices Now in Android #119 Getting Backstage in front of a shifting dev experience From natural language to UI tests: A deep dive into Journeys for Android Studio 用开发 Agent 理解 Agent 01:握着你的手摸索小龙虾 用开发 Agent 理解 Agent 02:数据即代码,代码即数据 Your AI Robo-Buddy Is Legally Required to Grass You Up Capturing token IDs during agentic interactions for better reinforcement learning Birgitta Boeckeler on Harness Engineering for AI Agents Eval-driven development: Lessons from evaluating GenAI at scale Flexible Authentication: Reimagining authentication for millions of users at Airbnb Bring one-handed gestures to your Wear OS app Preparing your app for broader memory limits Elevating app quality: Reducing memory usage and improving device migration Emulator control for adaptive app development Leverage Android skills and Gemma 4 in Android Studio Quail 4 Beyond the model: Engineering AI infra with scientific judgement The guest journey, updated in real time: extending Airbnb’s sequence recommender with Chronon Bring your Android game to the car screen today Land your apps on Googlebook with adaptive development Driving growth on Google Play: The next era of subscriptions Android Bench 2.0: Pushing the frontier with challenging long-horizon tasks Device Streaming and Android skills - available in Android CLI Beyond synthetic testing: Capturing and replaying real database workloads at Airbnb Introducing the Googlebook Linux Terminal Preview: A VM-Backed Environment for AI and Development Are Harnesses Worth the Cost? New ways to drive smarter discovery, deeper engagement, and business growth on Google Play Bringing our latest frontier model to Android Developers Android Developers Blog is moving Your Discord Messages Will Soon Be Discoverable on Google Search, but You Can Opt Out Why a 1B On-Device LLM Needs a Harness Google is redesigning Messages again, and heavy texters might hate it Agent Harness: Definition Android Dev Summit 2026 Gesture Controls With Only a Camera Почему мощная нейросеть может работать плохо: что такое харнесс и как он меняет возможности AI-агентов Google updates 8 Translate widgets with a clean, refined new design Henry Cavill’s Mustache Protocol Googlebook game is now ready to play Motorola’s new wireless Android Auto dongle is finally here, but it has a big price hike London hotel put admin login for booking website on whiteboard behind the front desk Motorola MA2 Adds a Phone-Switching Button to Wireless Android Auto Exclusive: Google’s hidden list of 134 apps that support Gemini task automation Claude’s Duet agent is a little neurotic and surprisingly helpful Google Messages’ big search redesign is now rolling out to beta testers I finally have Google Messages’ new pinned chats UI. Here’s how to get it on your phone now 5 AI agents you can actually use on Android in 2026 (and who should use them) Gmail’s Material 3 Expressive settings redesign is starting to reach Android users Tested in the wild: Best survival tech! Paul Bakaus: AMP is a pretty misunderstood project
- T+12h 110 篇 2026-11-19 09:40 Mobile: Unit Testing Fixing Duplicate Contacts in Android People app from Google Contacts and Google+ Writing your own Karma adapter Automatic Stubbing of Network Requests to Deflakify Automation Testing Running Android UI Test Suites on Firebase Test Lab Unit Testing vs Integration Tests Increased accuracy of aapt2 "keep" rules Google AMP’s Malte Ubl wants to make the mobile web better Integration Testing for Memory Leaks Travel, Parenting, and ADHD with Heidi Waterhouse Heidi Helen is a Designer/Developer/Business Owner Selectively running Android modularized unit tests on your CI server Better Android testing at Airbnb, Part 5 Better Android testing at Airbnb, Part 6 Better Android testing at Airbnb, Part 7 Tests Under the Magnifying Lens How we enable Airbnb team members to code like a mobile engineer Airbnb’s Page Performance Score on Android Making Airbnb’s Android app more accessible Spotify and Disney have everything-app dreams 173: Mocking and Unit Tests Between hyper-focus and burnout: Developing with ADHD How Dropbox leverages testing to maintain high level of trust at scale Gustav Söderström - How Spotify Thinks A product manager’s guide to adapting Android apps across devices Now in Android #119 Getting Backstage in front of a shifting dev experience From natural language to UI tests: A deep dive into Journeys for Android Studio 用开发 Agent 理解 Agent 01:握着你的手摸索小龙虾 用开发 Agent 理解 Agent 02:数据即代码,代码即数据 Your AI Robo-Buddy Is Legally Required to Grass You Up Capturing token IDs during agentic interactions for better reinforcement learning Birgitta Boeckeler on Harness Engineering for AI Agents Eval-driven development: Lessons from evaluating GenAI at scale Flexible Authentication: Reimagining authentication for millions of users at Airbnb Bring one-handed gestures to your Wear OS app Preparing your app for broader memory limits Elevating app quality: Reducing memory usage and improving device migration Emulator control for adaptive app development Leverage Android skills and Gemma 4 in Android Studio Quail 4 Beyond the model: Engineering AI infra with scientific judgement The guest journey, updated in real time: extending Airbnb’s sequence recommender with Chronon Bring your Android game to the car screen today Land your apps on Googlebook with adaptive development Driving growth on Google Play: The next era of subscriptions Android Bench 2.0: Pushing the frontier with challenging long-horizon tasks Device Streaming and Android skills - available in Android CLI Beyond synthetic testing: Capturing and replaying real database workloads at Airbnb Introducing the Googlebook Linux Terminal Preview: A VM-Backed Environment for AI and Development Are Harnesses Worth the Cost? New ways to drive smarter discovery, deeper engagement, and business growth on Google Play Bringing our latest frontier model to Android Developers Android Developers Blog is moving Your Discord Messages Will Soon Be Discoverable on Google Search, but You Can Opt Out Why a 1B On-Device LLM Needs a Harness Google is redesigning Messages again, and heavy texters might hate it Agent Harness: Definition Android Dev Summit 2026 Gesture Controls With Only a Camera Почему мощная нейросеть может работать плохо: что такое харнесс и как он меняет возможности AI-агентов Google updates 8 Translate widgets with a clean, refined new design Henry Cavill’s Mustache Protocol Googlebook game is now ready to play Motorola’s new wireless Android Auto dongle is finally here, but it has a big price hike London hotel put admin login for booking website on whiteboard behind the front desk Motorola MA2 Adds a Phone-Switching Button to Wireless Android Auto Exclusive: Google’s hidden list of 134 apps that support Gemini task automation Claude’s Duet agent is a little neurotic and surprisingly helpful Google Messages’ big search redesign is now rolling out to beta testers I finally have Google Messages’ new pinned chats UI. Here’s how to get it on your phone now 5 AI agents you can actually use on Android in 2026 (and who should use them) Gmail’s Material 3 Expressive settings redesign is starting to reach Android users Tested in the wild: Best survival tech! Paul Bakaus: AMP is a pretty misunderstood project
- T+24h 110 篇 2026-11-19 21:40 Mobile: Unit Testing Fixing Duplicate Contacts in Android People app from Google Contacts and Google+ Writing your own Karma adapter Automatic Stubbing of Network Requests to Deflakify Automation Testing Running Android UI Test Suites on Firebase Test Lab Unit Testing vs Integration Tests Increased accuracy of aapt2 "keep" rules Google AMP’s Malte Ubl wants to make the mobile web better Integration Testing for Memory Leaks Travel, Parenting, and ADHD with Heidi Waterhouse Heidi Helen is a Designer/Developer/Business Owner Selectively running Android modularized unit tests on your CI server Better Android testing at Airbnb, Part 5 Better Android testing at Airbnb, Part 6 Better Android testing at Airbnb, Part 7 Tests Under the Magnifying Lens How we enable Airbnb team members to code like a mobile engineer Airbnb’s Page Performance Score on Android Making Airbnb’s Android app more accessible Spotify and Disney have everything-app dreams 173: Mocking and Unit Tests Between hyper-focus and burnout: Developing with ADHD How Dropbox leverages testing to maintain high level of trust at scale Gustav Söderström - How Spotify Thinks A product manager’s guide to adapting Android apps across devices Now in Android #119 Getting Backstage in front of a shifting dev experience From natural language to UI tests: A deep dive into Journeys for Android Studio 用开发 Agent 理解 Agent 01:握着你的手摸索小龙虾 用开发 Agent 理解 Agent 02:数据即代码,代码即数据 Your AI Robo-Buddy Is Legally Required to Grass You Up Capturing token IDs during agentic interactions for better reinforcement learning Birgitta Boeckeler on Harness Engineering for AI Agents Eval-driven development: Lessons from evaluating GenAI at scale Flexible Authentication: Reimagining authentication for millions of users at Airbnb Bring one-handed gestures to your Wear OS app Preparing your app for broader memory limits Elevating app quality: Reducing memory usage and improving device migration Emulator control for adaptive app development Leverage Android skills and Gemma 4 in Android Studio Quail 4 Beyond the model: Engineering AI infra with scientific judgement The guest journey, updated in real time: extending Airbnb’s sequence recommender with Chronon Bring your Android game to the car screen today Land your apps on Googlebook with adaptive development Driving growth on Google Play: The next era of subscriptions Android Bench 2.0: Pushing the frontier with challenging long-horizon tasks Device Streaming and Android skills - available in Android CLI Beyond synthetic testing: Capturing and replaying real database workloads at Airbnb Introducing the Googlebook Linux Terminal Preview: A VM-Backed Environment for AI and Development Are Harnesses Worth the Cost? New ways to drive smarter discovery, deeper engagement, and business growth on Google Play Bringing our latest frontier model to Android Developers Android Developers Blog is moving Your Discord Messages Will Soon Be Discoverable on Google Search, but You Can Opt Out Why a 1B On-Device LLM Needs a Harness Google is redesigning Messages again, and heavy texters might hate it Agent Harness: Definition Android Dev Summit 2026 Gesture Controls With Only a Camera Почему мощная нейросеть может работать плохо: что такое харнесс и как он меняет возможности AI-агентов Google updates 8 Translate widgets with a clean, refined new design Henry Cavill’s Mustache Protocol Googlebook game is now ready to play Motorola’s new wireless Android Auto dongle is finally here, but it has a big price hike London hotel put admin login for booking website on whiteboard behind the front desk Motorola MA2 Adds a Phone-Switching Button to Wireless Android Auto Exclusive: Google’s hidden list of 134 apps that support Gemini task automation Claude’s Duet agent is a little neurotic and surprisingly helpful Google Messages’ big search redesign is now rolling out to beta testers I finally have Google Messages’ new pinned chats UI. Here’s how to get it on your phone now 5 AI agents you can actually use on Android in 2026 (and who should use them) Gmail’s Material 3 Expressive settings redesign is starting to reach Android users Tested in the wild: Best survival tech! Paul Bakaus: AMP is a pretty misunderstood project
Verify Card
| Type | Claim | Status | Basis |
|---|---|---|---|
| Opposition | diff | 对立方相似度 69% | |
| Opposition | diff | 对立方相似度 62% | |
| Source flip | diff | 同源 0 篇立场翻转 |
Conclusion
该事件汇聚 73 个来源,平均可信度 46%
Confidence: medium
Find Your Interest
- How Meta Built Forum With AI-Native Engineering
- The Roundup by ByteByteGo
- Introducing AI Code Agents: A TypeScript SDK to Solve Vendor Lock-in for Coding Agents
- wsf-xxxviii
- wsf-xxxvi
- wsf-xxxi
- JetBrains IDEs Transformed for the Agentic Era of Development – With Air
- The First AI Model That Scared Me Can’t Write a Sentence
- The Agentic Web
- Everyone Hates AI. Half of Them Use It.





