Gemini 2.5 Flash download Archives - Tax Heal

Gemini API Managed Agents: 3.6 Flash, hooks, and more

Ashwani Kumar — Tue, 04 Aug 2026 12:49:55 +0000

Gemini API Managed Agents: 3.6 Flash, hooks, and more

Managed Agents in Gemini API now default to Gemini 3.6 Flash. New environment hooks let you block, lint, or audit tool calls inside the sandbox. Also, we’ve added budget controls, scheduled triggers, and free tier access.

Managed Agents in Gemini API are getting environment hooks, model selection, and free tier access. These capabilities build on our previous release introducing background tasks and remote MCP server integration.

With managed agents in the Gemini Interactions API, a single API call coordinates, reasoning, code execution, package installation, file management, and web retrieval inside an isolated cloud sandbox.

If you’re using an AI coding assistant, drop this in your terminal to give it access to the Interactions API skill: npx skills add google-gemini/gemini-skills --skill gemini-interactions-api.

Below are examples using the @google/genai TypeScript/JavaScript SDK. For Python or cURL, check out the Antigravity agent documentation.

npm install @google/genai

Gemini 3.6 Flash is now the default

The antigravity-preview-05-2026 agent now runs Gemini 3.6 Flash by default. No code changes are required. Your next interaction picks it up automatically.

You can also explicitly select models by passing agent_config.model when creating an interaction or managed agent. Use Gemini 3.5 Flash-Lite for lower cost, or pin to your model of preference.

import { GoogleGenAI } from "@google/genai";

const client = new GoogleGenAI({});

const interaction = await client.interactions.create({
  agent: "antigravity-preview-05-2026",
  input: "Audit all dependencies in package.json, upgrade outdated packages, and verify the build by running npm test.",
  environment: "remote",
  agent_config: {
    type: "antigravity",
    model: "gemini-3.5-flash-lite",
  },
});

console.log(interaction.output_text);

Supported models include:

Gemini 3.6 Flash (gemini-3.6-flash, default): Balanced model for reasoning, coding, and tool use.
Gemini 3.5 Flash (gemini-3.5-flash): Previous generation for general agentic workflows.
Gemini 3.5 Flash-Lite (gemini-3.5-flash-lite): Lowest latency and cost on the Gemini 3.5 family.

Environment hooks: block, lint, and audit tool calls inside the sandbox

Environment hooks let you run your custom scripts before or after every tool call the agent makes inside its sandbox. Add a .agents/hooks.json into your environment and the runtime executes your handlers on pre_tool_execution or post_tool_execution events.

The matcher field supports regular expressions, allowing you to target multiple tools with | or catch everything with *:

{
  "security-gate": {
    "pre_tool_execution": [
      {
        "matcher": "code_execution|write_file",
        "hooks": [
          {
            "type": "command",
            "command": "python3 /.agents/hooks-scripts/gate.py",
            "timeout": 10
          }
        ]
      }
    ]
  },
  "auto-format": {
    "post_tool_execution": [
      {
        "matcher": "*",
        "hooks": [
          {
            "type": "command",
            "command": "python3 /.agents/hooks-scripts/auto_lint.py",
            "timeout": 15
          }
        ]
      }
    ]
  }
}

In this configuration:

The security-gate group runs gate.py before every code_execution or write_file call. If the script returns {"decision": "deny", "reason": "..."}, the tool call is skipped and the rejection reason is passed into the model’s context.
The auto-format group runs auto_lint.py after every tool finishes to enforce code styling.
Hooks also support http type handlers that POST directly to an external endpoint.

For complete HTTP hook definitions and failure handling semantics, refer to the hooks documentation.

Teams are already using hooks to build production-grade validation pipelines. For example, AI-native investment bank Offdeal uses post_tool_execution hooks to run automated image verification inside the remote sandbox.

“OffDeal is an AI-native investment bank, and Archie is the AI analyst our bankers use every day. A requirement for banker-ready decks is company logos: buyer tables, sponsor columns, tombstone grids, often 30+ logos in a single deck, every one of which must be the right company, the appropriate size and aspect ratio, contain the name, have a transparent background, and have a high contrast when placed on a white slide.

Before agent hooks, we couldn’t do this on Gemini’s managed agents: the sandbox is remote, so our validation code had nowhere to run. With hooks, a post_tool_execution hook triggers our pipeline inside the sandbox the moment Archie writes its company list, fetching candidates, enforcing pixel-level quality checks, verifying each logo with Gemini vision, and publishing a manifest of approved files that are the only images allowed into the deck.”

– Alston Lin, Founder & CTO of OffDeal

Cost control and automation features

Free tier availability

Managed agents are now available on free tier projects. Developers can experiment with agentic workflows using an API key from a project without active billing.

Budget controls

Because managed agents execute multi-turn autonomous loops, complex tasks can consume significant token budgets. To prevent runaway tasks, you can pass max_total_tokens inside agent_config to cap total consumption (input + output + thinking).

When the agent reaches the limit, execution safely pauses and the interaction returns status: "incomplete". The environment state is preserved, enabling you to continue where it stopped by passing previous_interaction_id with a fresh budget.

const interaction = await client.interactions.create({
  agent: "antigravity-preview-05-2026",
  input: "Audit all modules in this repo and generate a migration report.",
  agent_config: {
    type: "antigravity",
    max_total_tokens: 10000,
  },
  environment: "remote",
});

Scheduled execution with triggers

Automate recurring agent tasks with scheduled triggers. A trigger binds an agent, environment, prompt, and cron schedule into a persistent resource that fires without manual intervention. Each run reuses the same sandbox, so files persist across executions.

Environments API

The Environments API lets you list, inspect, and delete sandbox sessions from code. Recover environment IDs after a disconnect, or clean up sandboxes when your pipeline finishes instead of waiting for the 7-day TTL.

Get started with managed agents

These updates turn managed agents into cost-controlled, scheduled workers that operate autonomously inside real development environments without breaking your budget or requiring external orchestration.

Check out the Gemini Interactions API overview and the managed agents quickstart to explore custom agent definitions, environment configurations, network rules, and advanced streaming patterns.

The Google DeepMind update for Managed Agents in the Gemini API introduces a major shift toward native, platform-level agent orchestration. Instead of forcing developers to build complex custom middleware to manage autonomous loops, Google has integrated a secure runtime control layer right into their isolated cloud sandboxes. [1, 2]

Here are the key upgrades introduced in this release:

Gemini 3.6 Flash as the Default Model

Automatic upgrade: Managed agents now default to the new Gemini 3.6 Flash model, requiring zero code modifications.
Efficiency gains: The 3.6 Flash model reduces overall output token usage by 17%, and cuts token consumption by up to 65% on long-horizon software engineering tasks.
Manual selection: Developers can still override the default to choose Gemini 3.5 Flash or Gemini 3.5 Flash-Lite using agent_config. [2, 3, 4, 5]

Synchronous Environment Hooks

Tool intercepting: New synchronous hooks let you intercept code execution and file system tools inside the remote Linux sandbox.
Pre & post events: Features explicit pre_tool_execution and post_tool_execution event hooks.
Runtime guardrails: Allows you to block dangerous commands, lint code syntax, or audit tool actions using regex matching before they execute. [1, 2, 6, 7]

Production & Budget Controls

Runaway protection: Implements token budget limits via max_total_tokens parameters to halt endless, runaway autonomous loops.
State preservation: When budget caps are triggered, the agent pauses automatically without losing its active sandbox environment state.
Environments API: Provides new endpoints to inspect, list, and explicitly delete active sandboxed environments.
Scheduled triggers: Supports background execution and automated task scheduling without requiring active HTTP socket connections. [2, 7, 8, 9, 10]

Accessibility & Ecosystem Updates

Free tier access: Managed agents are now available on free tier projects, allowing you to test agentic workflows with a standard API key without active billing. [1, 9]
MCP Integration: Fully integrates with the remote Model Context Protocol (MCP) to seamlessly connect your custom tools and databases. [1, 2]
Interactions API GA: The underlying Interactions API has reached general availability, rendering it Google’s primary default schema for Gemini development. [10]

To learn how to implement these parameters in your code, check out the official Google Custom Agents Guide. [3]

If you are currently migrating your system, let me know:

What specific tools (e.g., Python execution, bash commands, web search) your agent uses?
If you need a code example showing how to structure the new environment hooks or budget parameters? [2, 6, 7]

. How to know if an image is AI generated or not

. Understanding the AI economy

. Apply now for the Google for Startups Gemini Startup Forum.

**********************************************************************

. Use Gemini in Google Forms to quickly create a new quiz

. Prevent accidental disclosures with new Reply All BCC warnings in Gmail

. Dilip Gavit Wins Gold & Mohammed Basil Wins Silver | Highlights | Commonwealth Games Glasgow 2026

. Neeraj Chopra is competing in the men’s javelin throw final at the 2026 Commonwealth Games in Glasgow

. Play Points members and comic book fans can score exclusive rewards with Google Play

. Google is signing the EU AI Act Code of Practice on Transparency of AI-Generated Content

. Gemini Robotics 2 brings whole body intelligence to robots

. NOAA and Google Cloud collaborate to advance weather forecasting.

. Google Workspace Weekly Recap – July 31, 2026

. Meet Qwen3.8-Max: A New Bar for Coding and Cowork

. 5 ways to host the ultimate dinner party with Google Search

. 5 ways AI Mode in Search helps you enjoy the real world

. These Google Trends show people really want to touch grass

. 2026-07-01 124 344 Codex Pets Meet The Team USEN YOUTUBE 1920×1080

. OpenAI’s New AI ASTRA Is Total Overkill (Ends The GPT Era)

. 3 new ways nonprofits can put AI to work

for more refer Gemini website click here

for more refer Artificial Intelligence website click here

How Gemini 3.6 Flash helped this dairy farmer run his business

Ashwani Kumar — Thu, 30 Jul 2026 11:30:38 +0000

How Gemini 3.6 Flash helped this dairy farmer run his business

#AIAgent #SmartFarming #GoogleGemini

Dairy farming is a highly complex biological system driven by unpredictable market forces. In this video, see how Paul, a Michigan dairy farmer uses Gemini 3.6 Flash in Google Antigravity to build an AI dashboard that automates daily performance tracking. Gemini Flash cuts token costs while combining vision and reasoning to automate farm financial calculations from photos alone, freeing up time for hands-on work without sacrificing operational insights.

#GoogleGemini #SmartFarming #AIAgent

Artificial Intelligence is transforming modern agriculture by shifting farming from manual, intuition-based labor to data-driven, precision operations. As highlighted in the video (0:00 – 2:30), AI allows farmers to manage highly complex biological systems—like a dairy farm—more efficiently.

How AI is Revolutionizing Farming:

Data Integration & Monitoring: Farmers can now collect hundreds of thousands of data points daily using sensors (such as cow collars) to track health, milk quality, feed, and weather (0:40 – 0:51).
Automated Decision-Making: AI “agents” can process this vast amount of data to perform complex calculations, such as financial performance or margin tracking, which would otherwise be mentally taxing (0:56 – 1:21).
Computer Vision: By analyzing photos of the farm, AI can identify metrics and provide operational insights that the farmer might not have explicitly input, effectively acting as an “AI dashboard” (1:46 – 1:58).
Efficiency and Cost Reduction: Models like Gemini 3.6 Flash allow farmers to build and run these agents without high token costs, making AI implementation consistent and sustainable for daily business management (1:39 – 1:43).

Broader Impact on Agriculture:

Beyond dairy, AI is impacting the industry through:

Precision Agriculture: Using machine learning to optimize irrigation, nutrient management, and planting/harvesting tasks to reduce waste and increase yields.
Livestock Management: AI-powered systems monitor animal health, predict disease outbreaks, and improve breeding efficiency.
Automated Labor: Robotics and AI are increasingly handling labor-intensive tasks like weeding and crop monitoring, reducing the reliance on manual labor.

Michigan dairy farmer Paul Windemuller automated his daily operations and tracking by building an AI system powered by Gemini 3.6 Flash inside Google Antigravity. Dubbed “The Farm Brain,” this localized, multi-agent AI framework processes vast amounts of agricultural data, allowing Paul to spend less time crunching numbers in the office and more time caring for his herd of 260 Holstein cows. [1, 2, 3, 4]

The Problem: Fragmented Data Silos

Modern automated farms generate massive continuous data streams. Paul’s facility collected metrics from 24/7 cow sensor collars, local weather stations, feed logs, and online milk quality databases. Because these data streams were trapped in separate software silos, Paul previously had to spend hours every morning manually downloading files, merging spreadsheets, and calculating profit margins. [1, 2, 5]

The Solution: Multi-Agent AI Workflow

Using Gemini 3.6 Flash’s multimodal and advanced reasoning features, Paul constructed a directory-based file-interface system that processes farm data without requiring fragile web APIs. When CSV files, digital invoices, or even smartphone photos of paper receipts are saved to a local folder, the AI extracts the metrics on-premises. [2, 6, 7]

Four specialized, coordinated AI agents manage the daily workload: [8]

Orchestrator Agent: Manages and oversees the overarching daily file processing workflow.
Ingestion Agents: Standardize disparate file formats, including feed logs and milking robot data exports.
Analysis Agent: Cross-references and evaluates how biological changes and local weather impacts affect production.
Reporting Agent: Generates daily briefings in clear, natural language, summarizing the farm’s exact health and performance metrics. [2, 5]

Maximizing Efficiency and Tracking Profitability

The AI specifically optimizes for a metric called Daily Static Variable Margin (SVM). Traditional farming calculations can be heavily skewed by volatile shifts in market milk and feed prices. By keeping market variables constant, the Gemini-powered system isolates true biological and operational efficiency, giving Paul reliable, actionable insights into how weather or health changes are affecting his business. [2]

Why Gemini 3.6 Flash?

Agentic workflows require continuous multi-step reasoning, making them historically too expensive for small businesses. Gemini 3.6 Flash significantly lowered operational costs for the farm due to a 17 percent reduction in output tokens and a lower cost per token compared to previous models, making continuous agent execution economically viable at a small-business scale. [2, 6, 9]

Would you like to learn more about how to build a local multi-agent AI workflow or explore other real-world applications of Gemini 3.6 Flash in agriculture?

. Switch to Android easily and bring your data with you.

. Commonwealth Games 2026 Highlights, July 28: Gulveer Singh, Harjinder Kaur win landmark silver medals

. Google’s Illyes Unsure On Shifting unavailable_after Dates

. Streamline collaboration in Google Docs with Gemini-powered comment workflows

. What Exactly Is USB-C? (The Real Reason Apple Lost)

. Generate and edit visuals with Gemini in Google Docs570. 3 Google updates from Galaxy Unpacked 2026

. Here’s how to ask Gemini Live for help with anything you see.

. 2026 earnings call: Remarks from our CEO

. Introducing selfie for sign-in: a new, easy way to access your Google Account

. Updated options to better view Google Calendar on large monitors

. Create and import scatter charts with multiple x-series in Google Sheets

. Google Pixel 11 Pro | The Next Obvious Move

for more refer Gemini website click here

for more refer Artificial Intelligence website click here

Introducing Gemini 3.6 Flash, 3.5 Flash-Lite, and 3.5 Flash Cyber

Ashwani Kumar — Mon, 27 Jul 2026 13:05:56 +0000

Introducing Gemini 3.6 Flash, 3.5 Flash-Lite, and 3.5 Flash Cyber

Our newest Gemini models deliver the efficiency, latency, and reliability to build AI agents at scale.

Developers and customers building production AI agents need higher token efficiency, lower latency, and more reliable performance. Our Flash series of models is built to meet the sweet spot of efficiency and quality to enable scaling agentic workflows. Building on Gemini 3.5 Flash, we’re introducing new Gemini models:

3.6 Flash: Our workhorse model that delivers better coding, knowledge work, and multimodal performance. According to the Artificial Analysis Index, it reduces output token usage by 17% compared to 3.5 Flash, and in some benchmarks like DeepSWE by Datacurve, we observe up to 65%, all at a lower cost per output token.
3.5 Flash-Lite: Our fastest, most cost-effective 3.5-class model, delivering 350 output tokens per second according to the Artificial Analysis Index, also significantly outperforming prior Flash-Lite generations in agentic workflows.
3.5 Flash Cyber in CodeMender: Successful cybersecurity applications require careful orchestration of a model alongside an agent infrastructure. We’re introducing a combination of a new, highly efficient, specialized cyber-focused model paired with our CodeMender code security agent that delivers competitive performance at the frontier.

Beyond today’s releases, Gemini 3.5 Pro is currently testing with partners and we plan to make it broadly available as soon as it’s ready. In parallel, our team is already focusing on building the next generation of models. We have started our most ambitious pre-training run yet, for Gemini 4, and are excited by the progress.

3.6 Flash: More efficient and better quality than 3.5 Flash

Gemini 3.6 Flash builds directly on developer and customer feedback from 3.5 Flash. 3.6 Flash not only delivers a step up in coding and knowledge work, but it does this while meaningfully improving token efficiency. For example, on the Artificial Analysis Index, we see 3.6 Flash consuming 17% fewer output tokens than 3.5 Flash. It also takes fewer reasoning steps and tool calls to accomplish multi-step workflows.

This enhanced efficiency is also combined with a lower price than 3.5 Flash. At $1.50/1M input tokens and $7.50/1M output tokens, 3.6 Flash reduces the overall cost per agentic task, making agents more cost-effective to build and run.

https://www.taxheal.com/wp-content/uploads/2026/07/token_efficiency.mp4

3.6 Flash shows better token efficiency and reduced verbosity than 3.5 Flash in an OSWorld verified task (API)

Even while being more efficient, 3.6 Flash sees performance gains compared to 3.5 Flash across use cases:

3.6 Flash delivers higher precision with fewer unwanted code edits and reduced execution loops, as seen in DeepSWE (49% vs. 37%), and shows significant improvement in ML Research, as seen in MLE Bench (63.9% vs. 49.7%).
It has improved computer use capabilities as seen in OSWorld-Verified (83.0% vs. 78.4%). Computer use is now a built-in client side tool via the Gemini API and Gemini Enterprise.
It outperforms 3.5 Flash in knowledge work, as shown by benchmarks like GDPval-AA v2 (1421 vs. 1349). Customers like Hebbia and Harvey have found it particularly capable at multimodal tasks like document parsing, chart and data analysis, and report drafting.

https://www.taxheal.com/wp-content/uploads/2026/07/token_efficiency-1.mp4

3.6 Flash, using Managed Agents on AIS, can help parse through and analyze financial data and transcripts more efficiently and accurately than 3.5 Flash (AIS)

https://www.taxheal.com/wp-content/uploads/2026/07/1139_DEMO_Gemini36_sxs_TK_4K_16x9_compressed_30mb.mp4

3.6 Flash executes code migrations, using multi-agent orchestration on AGY, with lower latency and higher quality than 3.5 Flash (AGY)

https://www.taxheal.com/wp-content/uploads/2026/07/Google_Demo_Lab_TextureExtractor_16x9_v05-_36endcard_1.mp4

3.6 Flash helps develop a photographic texture extractor for 3D workflows, using canvas (Gemini App)

https://www.taxheal.com/wp-content/uploads/2026/07/1140_DEMO_TLDRaw_ThemeStudio_JP_4K_16x9_compressed_30mb_2.mp4

3.6 Flash uses AGY and the tldraw offline editor to build interactive theme studios with its strong visual understanding skills (AGY).

Customers report 3.6 Flash is a step forward in both cost and quality, balancing token efficiency, accuracy, and speed across complex workflows and knowledge-based tasks:

Built with safety

3.6 Flash is shipping with enhanced Frontier Safety safeguards in the domains of Chemical, Biological, Radiological, and Nuclear (CBRN) and cyber offense misuses. These safeguards make the model substantially more resistant to jailbreaks. At the same time, the model has been trained to minimize refusals for beneficial uses.

For more information, see the 3.6 Flash model card.

3.5 Flash-Lite: Built to scale agentic workflows

Beyond Flash, we’re also releasing Gemini 3.5 Flash-Lite, designed for both low-latency tasks and tasks where high throughput is critical for developers workflows, like agentic search and document processing.

3.5 Flash-Lite is the fastest model in the 3.5 series. As measured by Artificial Analysis, it runs at 350 output tokens/s. Priced at $0.3/1M input tokens and $2.5/1M output tokens and with significantly better quality than 3.1 Flash-Lite, 3.5 Flash-Lite offers a strong price-to-performance ratio for developers and customers running high throughput production traffic.

https://www.taxheal.com/wp-content/uploads/2026/07/modelComparator-35FL-vs-3.5F-260720c.mp4

3.5 Flash-Lite enables efficient scaling for agentic systems. Across thinking levels, the model significantly outperforms 3.1 Flash-Lite. Depending on the workload, developers can configure the model to prioritize low-latency, low-cost execution for high-volume tasks with the minimal and low thinking levels, or engage higher thinking levels to process multi-step subagent workloads. The model now also has computer use as a built-in tool to reliably support these agentic tasks across surfaces.

It’s a significant step up in coding and agentic tasks as seen in Terminal-Bench 2.1 (54% vs 31%), long context as seen in GDM-MRCR v2 (72.2% vs. 60.1%), and real-world task execution as seen in GDPval-AA v2 (1140 vs. 642).

Introducing Gemini 3.6

In fact, on many agentic and coding evals, 3.5 Flash-Lite even outperforms 3 Flash, including on SWE-Bench Pro (54.2% vs. 49.6%) and OSWorld-Verified (74.0% vs. 65.1%), making it a faster & more capable option for workloads on both 2.5 and 3 Flash.

https://www.taxheal.com/wp-content/uploads/2026/07/Google_DemoLab_ExtractALot_16x9_sw_08_EndCard_01_1.mp4

3.5 Flash-Lite extracts product features from a massive e-commerce dataset and synthesizes it.

https://www.taxheal.com/wp-content/uploads/2026/07/Google_Demo_Lab_ReceiptScanner_Developer_16x9_endcard-35_260721a.mp4

3.5 Flash-Lite can scale receipt translation and summarization with its multimodal understanding.

https://www.taxheal.com/wp-content/uploads/2026/07/Google_DemoLab_AgenticPuzzler_16x19_08_EndCard_01.mp4

3.5 Flash-Lite builds a game by instantly generating and iterating through multiple options.

Introducing Gemini 3.6

For more information about the model, see the 3.5 Flash-Lite model card.

3.5 Flash Cyber in CodeMender: finding and fixing vulnerabilities efficiently

AI models have become capable of finding security vulnerabilities faster than current systems can fix them. Tackling this growing threat requires an approach to securing software that is highly capable and efficient.

Flash’s performance and efficiency makes it an ideal foundation to detect, validate, and patch code security issues at scale. Gemini 3.5 Flash Cyber is built on top of 3.5 Flash, and fine-tuned for finding and fixing cybersecurity vulnerabilities at a lower price per token than larger models.

Within CodeMender, which uses multiple 3.5 Flash Cyber agents working together to produce a single combined report, 3.5 Flash Cyber reaches competitive performance at the frontier on the popular benchmark CyberGym.

Introducing Gemini 3.6

Given the dual-use nature of this technology, we have taken an intentional approach to deploying 3.5 Flash Cyber. The model will be exclusively available to governments and trusted partners via CodeMender soon as part of a limited-access pilot program. This will give frontline defenders a head start in finding and fixing critical vulnerabilities before they can be exploited, while mitigating against broader misuse.

3.6 Flash and 3.5 Flash-Lite: Get started today

3.6 Flash and 3.5 Flash-Lite are available starting today:

For developers in the Gemini API via Google AI Studio and Android Studio. 3.6 Flash is also available in Google Antigravity. Get started with the Developer Guide.
For enterprises in Gemini Enterprise Agent Platform. 3.6 Flash is also available in the Gemini Enterprise app.
For everyone via the Gemini app. 3.5 Flash-Lite is also rolling out in Google Search.

As you start building with 3.6 Flash and 3.5 Flash-Lite, we welcome your feedback to improve future Gemini models and look forward to releasing 3.5 Pro soon.