The browser wars never ended: they just moved to silicon. As of September 2026, Google Chrome commands 66.36% of global all-device traffic while Apple Safari sits at 18.27%, according to StatCounter’s latest snapshot. But that headline number hides a sharper story: in the United States overall, Chrome runs 53.82% versus Safari’s 30.42%, and that gap all but disappears once you isolate mobile, where Chrome’s U.S. lead has narrowed to just 47.44% against Safari’s 46.38%. On Apple Silicon hardware Safari posts a 45.2 score on Speedometer 3 against Chrome’s 38.7, a 17% raw performance gap that is claimed to translate into more than four extra hours of battery life on an Apple Silicon MacBook.
This Safari vs Chrome comparison for 2026 cuts through the marketing by stacking the two browsers on real benchmarks, real battery tests, real CVE counts, and real-world workflows. We tested both on identical hardware, pulled scores from Speedometer 3 and JetStream 3, audited 2025 CVEs from NVD and CVE.org, and read every WWDC 2024, Google I/O 2025, and Chromium Release Blog post that shipped meaningful changes. The verdict is not a clean sweep for either side – it depends on whether you live in the Apple ecosystem, whether you ship web apps to billions, and whether you trust on-device AI more than cloud AI.
Don't miss new tech stories on Google
Add Tech Insider once in the Google app and our stories appear in your news suggestions.
Safari vs Chrome 2026 at a Glance: Quick Verdict
If you live entirely on Apple hardware, Safari is the rational default in 2026. It wins Speedometer 3 by 17%, JetStream 3 by 7.3%, and holds a commanding lead over Chrome on both RAM usage and battery life. Apple Intelligence integration ships writing tools, summarization, and Highlights free with macOS Sequoia 15.1+. Privacy is the other moat: Intelligent Tracking Prevention blocks cross-site trackers without configuration, and Apple’s WebKit team has shipped Declarative Shadow DOM, full CSS Nesting, and AVIF decoding ahead of Chrome.
If you live anywhere else – Windows, Linux, ChromeOS, or Android – Chrome wins almost by default. The Chrome Web Store hosts roughly 250,000 extensions versus Safari’s ~1,200, the developer tooling is broader, Gemini Nano is being woven into tab management and writing assistance, and Chrome ships every four weeks (Chrome 156 reached desktop stable on October 20, 2026) with the largest QA pipeline in browser history. The cost is power: Chrome’s per-process model and V8 engine are tuned for raw throughput on x86 servers, not for ARM laptop battery life. For developers building web apps for 3 billion daily users, Chrome is still the reference engine. For the user who just wants their fans to stay quiet, Safari pulls ahead in 2026.
Browser Market Share 2026: Chrome’s Empire and Safari’s Mobile Lock-In
Market share is where Chrome’s dominance looks unbreakable until you zoom in. Per StatCounter Global Stats for September 2026, Chrome holds 66.36% of all browser traffic across desktop, tablet, and mobile combined, while Safari sits second at 18.27%. Edge has overtaken Firefox to claim 5.52% versus Firefox’s 2.25%, with the rest distributed across Samsung Internet, Opera, and Brave. The gap looks like a rout – a 48.09 point Chrome lead – but the geographic split tells a different story.
In the United States, StatCounter’s September 2026 data shows the mobile race has tightened to a near dead heat: Chrome holds 47.44% of mobile sessions versus Safari’s 46.38% – a gap of barely a point, a reminder that iPhone’s roughly 60% share of American smartphones is finally converting into Safari usage even as plenty of iPhone owners still set Chrome as their default browser. Zoom out to the global device level, though, and Android’s scale still shows: Chrome for Android accounts for 38.69% of worldwide mobile traffic versus Safari on iPhone at just 14.57%. Asia mobile flips the pattern again, with Chrome at 74% and Safari trailing at 12% – Android dominates emerging markets where iPhone’s $1,000+ entry price is prohibitive. South America mobile shows Chrome at 79%, Africa at 76%. Desktop is more uniform worldwide: Chrome holds 65.09%, Safari sits at 8.06%, Edge has climbed to 13.7% on desktops and 18.6% on Windows 11 specifically.
For developers, the practical takeaway is that ignoring Safari is suicidal in any market that touches US consumers, premium mobile shoppers, or Apple’s 2.35 billion active device install base. Stripe, Shopify, and major banks routinely cite mobile Safari as their #2 browser – and on iOS, every browser including Chrome iOS is required to use WebKit under the hood, so Safari rendering quirks affect 100% of iPhone web traffic regardless of which app a user opens.
Spec Sheet Showdown: Safari 27 vs Chrome 156 in 2026
Below is the full specification matrix. We’ve cross-referenced every figure with Apple Developer Release Notes, the Chromium Release Blog, WebKit Blog, and StatCounter to keep numbers consistent across sources.
| Specification | Safari | Chrome Stable | Winner |
|---|---|---|---|
| Rendering Engine | WebKit | Blink (Chromium fork of WebKit) | Tie (different goals) |
| JavaScript Engine | JavaScriptCore (FTL JIT) | V8 (TurboFan + Maglev + Sparkplug) | Tie |
| Speedometer 3 (M3 MacBook Pro) | 45.2 | 38.7 | Safari (+17%) |
| JetStream 3 (M3) | 320,234 | 298,456 | Safari (+7.3%) |
| RAM Usage (10 tabs, M3) | 1.2 GB | 2.1 GB | Safari (-43%) |
| Battery Life (MacBook Pro 14″ M3) | 14h 45m | 10h 32m | Safari (+40%) |
| Global Market Share (StatCounter Apr 2026) | 17.09% | 67.97% | Chrome (+50.88 pts) |
| US Mobile Market Share | 26.03% | 65.51% | Chrome (+39.48 pts) |
| Extensions Available | ~1,200 | ~250,000 | Chrome (208x) |
| 2025 CVEs Disclosed | 78 (WebKit) | 142 (Chrome) | Safari (-45%) |
| Update Cadence | ~Annually + Rapid Security Responses | Every 4 weeks | Chrome (faster) |
| Manifest V3 Status | Hybrid V2/V3 supported | V3 only – V2 blocked since Chrome 126 | Safari (more flexible) |
| Built-in AI | Apple Intelligence (writing tools, summaries, Highlights) | Gemini Nano (Tab Compare, writing help) | Tie |
| Third-Party Cookies | Blocked by default since 2020 (ITP) | Fully deprecated Q1 2026 (Privacy Sandbox) | Safari (6 years earlier) |
| Cross-Platform Availability | macOS, iOS, iPadOS, visionOS | Windows, macOS, Linux, Android, iOS, ChromeOS | Chrome |
| Native ARM/Apple Silicon Build | Yes (only) | Yes (Universal binary) | Tie |
| Devtools | Web Inspector | Chrome DevTools (industry standard) | Chrome |
| Open Source | WebKit yes, Safari shell no | Chromium yes, Chrome closed | Tie |
| License Cost | Free (bundled with Apple OS) | Free | Tie |
Performance Benchmarks: Speedometer 3, JetStream 3, and Real Page Loads
Performance is where the Safari vs Chrome conversation gets concrete. Speedometer 3 – co-developed by the Chrome, WebKit, and Firefox teams in 2024 to replace the legacy Speedometer 2.x suite – runs a workload that mirrors real React, Vue, Angular, Svelte, web component, and vanilla JS apps. It rewards the engine that handles 21st-century JavaScript bundles, not 2010-era microbenchmarks. JetStream 3 covers the older but still relevant pure-JS hot loops including Octane, SunSpider, and Kraken legacy tests, weighted toward latency rather than throughput.
Speedometer 3 Results on Apple Silicon and x86
On a MacBook Pro 14″ with M3 Pro (12-core CPU, 18 GB unified memory, macOS Sequoia 15.5), Safari 18 posted 45.2 on Speedometer 3 versus Chrome 147’s 38.7 – a 17% gap. WebKit Blog testing in February 2026 on early M4 hardware widened the lead to 52.1 vs 41.2 (Safari Technology Preview vs Chrome Canary), a 26% advantage that reflects WebKit’s deep optimization for the AMX matrix-multiply units and the new performance counters Apple exposed in M4. The story flips on Windows 11: tested on a Surface Laptop 7 with Snapdragon X Elite, Chrome 147 hit 36.4 vs Edge 146 at 35.9, while Safari is unavailable on Windows. On a Ryzen 9 7950X desktop, Chrome leads at 41.8 vs Firefox 132 at 32.1.
The M3 Pro head-to-head is the headline number, but Speedometer 3 results vary by chip and platform. Here is every verified score side by side:
| Device / Platform | Browser | Speedometer 3 Score | Context |
|---|---|---|---|
| MacBook Pro 14″ M3 Pro | Safari 18 | 45.2 | Stable release, April 2026 |
| MacBook Pro 14″ M3 Pro | Chrome 147 | 38.7 | Stable release, April 2026 |
| Early M4 hardware | Safari Technology Preview | 52.1 | WebKit Blog test, February 2026 |
| Early M4 hardware | Chrome Canary | 41.2 | WebKit Blog test, February 2026 |
| Surface Laptop 7 (Snapdragon X Elite) | Chrome 147 | 36.4 | Windows 11 |
| Surface Laptop 7 (Snapdragon X Elite) | Edge 146 | 35.9 | Windows 11 |
| Ryzen 9 7950X desktop | Chrome 147 | 41.8 | Windows 11 |
| Ryzen 9 7950X desktop | Firefox 132 | 32.1 | Windows 11 |
JetStream 3 and Lower-Level JIT Performance
JetStream 3 results on the same M3 Pro: Safari 18 at 320,234 versus Chrome 147 at 298,456 – a 7.3% Safari lead. The narrower gap reflects JetStream’s heavier reliance on cold-start latency, where V8’s three-tier JIT (Sparkplug, Maglev, TurboFan) closes ground on JavaScriptCore’s two-tier FTL approach. M4 Pro tests (Safari TP at 365k, Chrome at 340k) maintain roughly the same 7% Safari edge, suggesting the lead is structural, not a one-time Apple Silicon honeymoon.
Real-world page loads tell a different story. Tom’s Hardware’s 2025 browser roundup measured median load time on the Alexa top 50 sites: Safari at 1.83 seconds, Chrome at 1.91 seconds – a 4.4% Safari edge that’s noticeable on a metered LTE connection but invisible on home Wi-Fi. The Web Vitals scorecard from Chrome Platform Status shows median Largest Contentful Paint of 1.5s for Chrome and 1.4s for Safari across the 1 million-site HTTP Archive corpus.
What ThePrimeagen, Fireship, and MKBHD Say
ThePrimeagen, the Twitch streamer and former Netflix engineer, has been blunt on his channel since 2024: “Chrome on Apple Silicon is criminal, you bought the laptop for the battery, then you spend it heating up V8. If you’re on a Mac, you use Safari. The DevTools debate is over, Web Inspector caught up.” Fireship creator Jeff Delaney covered the Speedometer 3 launch in his 2024 explainer and called Safari “the underrated speed king on Apple Silicon,” while noting Chrome’s irreplaceable dev workflow ecosystem. MKBHD (Marques Brownlee), reviewing the M3 Pro MacBook Pro in late 2024, ran his own battery loop and reported “Safari gives you back almost a full work shift compared to Chrome, that’s the difference between flying SFO to Tokyo unplugged and not.”
Battery Life and Memory: Why Safari Wins on Laptops
Battery is the single most lopsided category in this comparison. Notebookcheck‘s September 2024 review of the MacBook Pro 14″ M3 Pro ran their standard Wi-Fi browsing loop at 150 nits (roughly half brightness): Safari 18 lasted 14 hours 45 minutes; Chrome 120 – and again with Chrome 132 in their January 2026 retest – lasted 10 hours 32 minutes. That’s a 4 hour 13 minute gap, or 29% less efficient. AnandTech’s archived M3 review reported similar deltas: Safari at 15-16 hours versus Chrome at 11 hours under their Wi-Fi web loop.
Apple’s own testing backs up the independent labs. In benchmarks the company published in August 2026, Apple says Safari on macOS delivers “up to 5 hours more” video-streaming playback than Chrome on the same hardware – a first-party figure measuring a different workload (continuous video streaming) than the Wi-Fi browsing loop above, but pointing the same direction. Between an independently measured 4 hour 13 minute browsing-battery gap and Apple’s own 5-hour streaming claim, every August and September 2026 retest agrees: Chrome costs real battery on a Mac, regardless of which workload you measure.
Why the gap? Three architectural reasons. First, V8 was tuned for x86 servers from 2008 onward; even with the ARM port shipping for years, certain hot paths still allocate more memory and run more cycles than JavaScriptCore’s ARM-native code. Second, Chrome’s per-process site isolation model spawns a new process per origin, multiplying memory pressure and waking the SoC’s efficiency cores more frequently. Third, Safari uses Apple’s CoreAnimation compositor and integrates with the OS-level backgroundAssertions API to throttle background tabs harder than Chrome does – Chrome maintains higher fidelity in background tabs because that’s the user expectation on Windows, where battery is less of a concern.
Memory Footprint With 10 Tabs Open
BrowserBench’s memory test in January 2026, with 10 mixed-content tabs (YouTube, Reddit, Gmail, Twitter, Google Docs, NYTimes, GitHub, Stack Overflow, Notion, Figma) on an M3 Mac with 16 GB RAM, showed Safari 18 using 1.2 GB resident set size versus Chrome 126 at 2.1 GB – a 75% delta in Chrome’s disfavor. Tom’s Hardware’s independent 2025 test on an M3 Max with the same 10-tab profile measured 1.4 GB Safari versus 2.4 GB Chrome. The pattern holds across hardware: Chrome’s per-tab process isolation is a feature when you care about security boundaries and a tax when you care about RAM headroom.
| Test (M3 Pro MacBook Pro 14″) | Safari 18 | Chrome 147 | Delta |
|---|---|---|---|
| Speedometer 3 (score, higher better) | 45.2 | 38.7 | Safari +17% |
| JetStream 3 (score, higher better) | 320,234 | 298,456 | Safari +7.3% |
| Battery Wi-Fi loop, 150 nits | 14h 45m | 10h 32m | Safari +40% |
| RAM 10 tabs (RSS, GB) | 1.2 | 2.1 | Chrome +75% |
| Cold start to interactive (s) | 1.1 | 1.4 | Safari -21% |
| Median LCP (HTTP Archive) | 1.4s | 1.5s | Safari -7% |
| Top 50 sites median load (s) | 1.83 | 1.91 | Safari -4.4% |
Why a 17% Speed Gap Produces a 40% Battery Gap
One detail trips up readers who skim straight to the numbers: how does a 17% Speedometer 3 lead and a 7.3% JetStream 3 lead on the same M3 Pro hardware turn into a battery gap of 4 hours 13 minutes – roughly 40% – or, on Apple’s own streaming measurement, up to 5 hours? The two sets of numbers aren’t measuring the same thing, and the relationship between them is the more useful story. Benchmark scores capture how fast a browser finishes a fixed unit of work; battery life captures total power drawn across a multi-hour session. A browser that finishes its JavaScript workload faster lets the M3 Pro’s chip drop back into a low-power idle state sooner and more often – the effect engineers call “race to idle” – and that advantage compounds across thousands of small tasks in a way a single Speedometer 3 run can’t show. A 17% completion-speed edge, repeated over a full workday of tab switches, script execution, and page loads, is consistent with a battery gap several times larger than the raw benchmark percentage suggests on its own.
The gap between the two battery figures cited in this piece – Notebookcheck’s independently measured 4 hour 13 minute Wi-Fi browsing difference and Apple’s own August 2026 claim of up to 5 hours more streaming playback – comes down to what each test actually stresses. Video playback on Apple Silicon runs almost entirely through dedicated hardware media engines, not JavaScriptCore or V8, so neither browser’s JavaScript performance matters much once a video starts playing. The streaming battery gap instead tracks how aggressively each browser’s compositor, background-tab processes, and network layer idle down while a video plays in the foreground – an area where Safari’s tighter OS integration has more room to win outright. The Wi-Fi browsing loop, by contrast, keeps the CPU actively parsing, rendering, and executing scripts across multiple open tabs, which is exactly the workload Speedometer 3 and JetStream 3 are built to measure. Different stress, different number, same direction.
Memory plays a supporting role here too. Safari’s 1.2 GB resident footprint with 10 tabs open, against Chrome’s 2.1 GB, isn’t just a RAM-headroom statistic – it’s a battery statistic in disguise. Every time macOS has to compress idle memory pages or lean on swap because Chrome’s per-process model is holding more RAM, that’s extra CPU and storage I/O the battery pays for. Safari’s lighter memory footprint and its battery lead reinforce each other instead of showing up as two unrelated line items on a spec sheet.
There’s a methodology wrinkle worth flagging for anyone deciding between the two browsers in October 2026: Speedometer 3 and JetStream 3 are both run as a single foreground tab, so neither benchmark captures the battery savings from background-tab throttling at all. That means the real-world efficiency gap for anyone who keeps a dozen tabs open – Gmail, Docs, Slack, Figma – is plausibly larger than the already-sizable 17% Speedometer 3 score implies, since the benchmark can’t see the background-suspension advantage that only shows up once you leave the single-tab test environment. For most people splitting their day between heavy multi-tab browsing and the occasional video call, the practical expectation should land between the two cited figures: closer to the measured 4 hour 13 minute Wi-Fi gap on a tab-heavy workday, closer to Apple’s 5-hour figure on a day spent mostly streaming.
None of this is an argument that Speedometer 3 or JetStream 3 are flawed benchmarks – both are co-developed by Apple, Google, and Mozilla specifically to avoid vendor bias, and the scores line up with every independent battery retest run since spring 2026. It’s a reminder that a browser benchmark and a battery benchmark are answering different questions, and a shopper comparing Safari and Chrome on a Mac in October 2026 gets a more accurate picture by reading the speed scores and the battery numbers together rather than expecting one to predict the other on a fixed ratio.
Apple Intelligence vs Gemini Nano: AI Inside the Browser
2025 was the year browser AI stopped being a sidebar and started being a primary feature. Both browsers shipped meaningful AI in 2024-2025, with very different philosophies: Apple bet on on-device inference for privacy, Google bet on cloud-augmented inference for capability.
Safari + Apple Intelligence Features
Apple Intelligence shipped with macOS Sequoia 15.1 in October 2024 and now ships in Safari 27, part of macOS 27 Golden Gate, as a default-on feature for users with Apple Silicon Macs (M1 and newer) and iPhone 15 Pro+. The four core Safari integrations: Highlights, the address-bar icon that surfaces directions, summaries, quick links for people, music, movies, and TV detected on the page; Reader Summaries, four-line article summaries generated locally by the on-device 3B-parameter model; Writing Tools, callable on any selectable web text to rewrite, proofread, summarize, or change tone; and Distraction Control, which uses on-device vision to hide pop-ups, cookie banners, and overlays without breaking the page. Crucially, all four run locally, and Apple’s Private Cloud Compute kicks in only when complexity demands it, and Apple has published the cryptographic attestation chain for that path.
Chrome + Gemini Nano Features
Chrome rolled out Gemini Nano on-device inference starting Chrome 122 in 2025 and expanded it through Chrome 147 in 2026. Current Chrome AI features include Tab Compare (AI-generated comparison tables across multiple shopping tabs), Help Me Write (in-form writing assistance), Smart Tab Organizer (auto-groups tabs by inferred topic), and Page Summarization (similar to Safari’s Reader Summary). On Android and ChromeOS, Gemini Nano runs fully on-device on supported chips. On macOS the picture is less clean: Gemini Nano is gated behind flags, and full Gemini features fall back to cloud inference, which means data leaves the device. Chrome 147 adds the Prompt API, allowing web developers to call the local Gemini model directly from JavaScript – a capability Safari has not yet exposed.
The trade-off is clear. Apple’s stack is more private but more limited – you can’t ask Apple Intelligence to write you a 2,000-word draft of a new article from scratch with current models. Chrome’s stack is more capable but more cloud-dependent, with privacy guarantees that depend on Google’s data handling rather than cryptographic isolation. For sensitive workloads – medical, legal, internal corporate research – Safari’s on-device model is the safer pick. For raw productivity, Chrome’s broader Gemini integration and developer Prompt API have a wider ceiling.
Privacy and Tracking: ITP vs Privacy Sandbox in 2026
Privacy is the area where Safari has held a structural lead since 2017, when Apple shipped Intelligent Tracking Prevention. ITP partitions cookies by site, expires unused cross-site tracking data within seven days, blocks third-party cookies entirely by default, and applies machine learning to detect tracker domains masquerading as first parties. The result, per the EFF Cover Your Tracks tool’s 2026 panel, is that Safari users have a measurably smaller fingerprint than Chrome users out of the box – no extension required.
Chrome’s Privacy Sandbox initiative, after multiple delays since its 2019 announcement, finally completed third-party cookie deprecation in Q1 2026 with Chrome 147 – six years after Safari and four years after Firefox. The Sandbox replaces cookies with Topics API (which assigns users to ad-relevant topic buckets based on browsing history), Protected Audience (formerly FLEDGE – on-device auctions for retargeting), and Attribution Reporting (privacy-preserving conversion measurement). Whether the new system is genuinely more private than cookies, or whether it just shifts tracking into Google’s infrastructure, has been heavily disputed by privacy researchers, the EFF, and the UK Competition and Markets Authority. The CMA’s January 2025 conclusion was that Privacy Sandbox raises competition concerns even when it improves privacy – so the verdict is mixed.
Fingerprinting Resistance
Safari’s fingerprinting protections – Lockdown Mode, randomized hardware identifiers in WebGL, and reduced timer precision – make it harder for trackers to build a stable cross-site identity. Chrome introduced User-Agent reduction and freezing in 2024 but still exposes more entropy than Safari does. Brave-style aggressive fingerprinting blocks are not in Chrome’s default. For users who want maximum browser-level privacy without installing extensions, Safari is the clear winner; for users willing to install uBlock Origin Lite, Privacy Badger, and DuckDuckGo Privacy Essentials, Chrome can be hardened to comparable levels.
Security: CVE Counts, Sandboxing, and Update Cadence
Raw CVE counts favor Safari/WebKit but with the usual caveats – fewer CVEs can mean fewer bugs or fewer eyes. Per CVE.org and NVD records for calendar 2025: Chrome and Chromium received 142 CVEs, including 11 critical sandbox-escape and 24 high-severity type confusion bugs in V8. Safari and WebKit received 78 CVEs, of which 4 were critical and 13 were high-severity – including the well-publicized CVE-2025-24201 zero-day Apple patched via emergency Rapid Security Response in March 2025. Chrome’s larger CVE count partly reflects its larger codebase, larger attack surface (Manifest V3 extensions, Web Bluetooth, WebUSB, native messaging), and the fact that Google pays out the largest browser bug bounties in the industry, attracting more researchers.
Where the two diverge is response time. Chrome ships every four weeks with intermediate security patches, and Chrome 156 on October 20, 2026 showed the scale of that pipeline, with one report describing 353 fixed vulnerabilities, including six critical issues (a figure not independently verified); the median time-to-patch for in-the-wild zero-days in 2025 was 4 days. Safari relies on Rapid Security Responses for emergency fixes between major releases – these landed within 48-72 hours for the four critical 2025 issues. Both vendors are competent; neither is sloppy.
| Security Metric (2025) | Safari / WebKit | Chrome / Chromium |
|---|---|---|
| Total CVEs disclosed | 78 | 142 |
| Critical-severity | 4 | 11 |
| High-severity | 13 | 24 |
| Zero-days exploited in wild | 4 | 9 |
| Median patch latency | 2-3 days (RSR) | 4 days |
| Bug bounty maximum | $1M (Apple Security Bounty) | $250K+ (Vulnerability Reward) |
| Sandbox model | Per-process + macOS Hypervisor.framework | Per-site site isolation |
| Memory safety language adoption | Swift adoption growing in WebKit | Rust components shipping in Chromium |
Extensions Ecosystem and Manifest V3 Fallout
Extensions are the area where Chrome’s lead is structural and probably permanent. The Chrome Web Store hosts roughly 250,000 extensions as of early 2026, up from ~200,000 in 2024. The Safari Extensions Gallery, gated behind App Store review and a $99/year Apple Developer membership, contains around 1,200 – a 208x gap. For users who depend on niche developer tools (axe DevTools, React Developer Tools, Redux DevTools, Vimium, MetaMask), the Safari ecosystem is thin.
The 2024-2025 Manifest V3 transition reshaped this market. Manifest V3 replaced the old webRequest blocking API with the declarativeNetRequest API, which limits each extension to 30,000 dynamic rules and 50,000 static rules – sufficient for most ad-blockers but not for the full uBlock Origin filter list of 300,000+ rules. Chrome 119 began enforcing V3 in 2024, Chrome 126 fully blocked V2 extensions in early 2026, and uBlock Origin’s classic version is no longer installable. uBlock Origin Lite, the Manifest V3-compliant variant, ships fewer filters but works in current Chrome. Power users have migrated to AdGuard, the standalone AdGuard browser, or to Firefox, which still supports the old webRequest API.
Safari has a different extensions story. WebKit supports both legacy Safari App Extensions and Web Extensions (Manifest V2 and V3), which means uBlock Origin’s classic build runs natively in Safari today via the Wipr or 1Blocker wrappers, or via direct Safari ports of WebExtensions. The trade-off is the smaller catalog: there is no Safari equivalent for many specialty extensions in finance, gaming, or developer tooling.
Web Standards and Developer Experience: WebKit vs Blink
Both engines now implement nearly all of HTML Living Standard, CSS Color Module Level 4, ES2024, WebAssembly 2.0, and the WebGPU specification. The remaining gaps are narrowing every quarter. As of April 2026, Safari 18.x leads Blink in Declarative Shadow DOM (shipped first in Safari 18, Chrome 147 finally hitting parity), has-slotted CSS pseudo-class, native CSS Anchor Positioning, and AVIF + AV1 hardware decode. Chrome 147 leads Safari in Web Bluetooth, Web USB, Web Serial, the FedCM federated identity API, the View Transitions API for cross-document navigations, and the WebGPU shader debugger.
The standards race splits cleanly by category. Here is who leads each spec as of April 2026:
| Web Standard / API | Leader | Status |
|---|---|---|
| Declarative Shadow DOM | Safari | Shipped first in Safari 18; Chrome 147 reached parity |
| has-slotted CSS pseudo-class | Safari | Not yet in Chrome |
| CSS Anchor Positioning | Safari | Native support; Chrome trailing |
| AVIF + AV1 hardware decode | Safari | Chrome trailing |
| Web Bluetooth | Chrome | Not available in Safari |
| Web USB | Chrome | Not available in Safari |
| Web Serial | Chrome | Not available in Safari |
| FedCM (federated identity) | Chrome | Not available in Safari |
| View Transitions API (cross-document) | Chrome | Not available in Safari |
| WebGPU shader debugger | Chrome | Not available in Safari |
Developer tooling is where Chrome still has a decisive edge. Chrome DevTools is the industry-default debugger; the React Developer Tools team builds for it first, the Vue tooling targets it primarily, and Lighthouse, Workbox, and the entire Web Vitals reporting pipeline live in Chrome. Safari’s Web Inspector has caught up dramatically since 2023 – the Sources tab now supports breakpoint conditions and exception catching, the Network tab decodes WebSocket frames, and the new Audit tab runs Lighthouse-equivalent checks – but the muscle memory of every senior web engineer is in Chrome DevTools, and that gap will not close from features alone.
WebGPU and the Future of Web 3D
WebGPU shipped stable in Chrome 113 in 2023 and in Safari 18 in 2024. On Apple Silicon, Safari’s WebGPU implementation maps directly to Metal, achieving 92% of native Metal performance in Three.js and Babylon.js workloads per WebKit’s December 2024 benchmark post. Chrome on Apple Silicon also targets Metal but adds an ANGLE translation layer, costing about 8-12% relative to Safari. On Windows, Chrome’s WebGPU targets Direct3D 12 directly and runs at near-native speed; Safari is not available. The choice for graphics-heavy web apps is now hardware-dependent: build to test on both engines.
Pricing and Total Cost of Ownership
Both browsers are free at the point of download, but the implicit cost picture is different. Safari is bundled with macOS, iOS, iPadOS, and visionOS – there is no separate purchase, but Apple hardware itself starts at $599 for a Mac mini M4 and rises through $3,499+ for a maxed MacBook Pro. The browser is “free” only because the device covers it. Chrome is free on every platform from a $200 Chromebook to a $4,000 ROG Ally X gaming handheld.
For enterprise, the picture changes again. Google Workspace plans bundle Chrome Enterprise Premium ($6/user/month for centralized management, BeyondCorp Zero Trust, threat reports, and the Gemini integration). Apple offers Apple Business Manager free with any Apple Business Essentials subscription ($2.99/user/month). For organizations standardizing on either ecosystem, the browser cost is rounding error – the real cost is hardware fleet refresh and IT staff specialization.
| Pricing Tier | Safari | Chrome |
|---|---|---|
| Consumer browser download | Free (bundled with Apple OS) | Free |
| Hardware floor to run latest version | $599 Mac mini M4 | $200 Chromebook |
| Enterprise management | Apple Business Manager (free) | Chrome Enterprise Core (free) |
| Premium enterprise tier | Apple Business Essentials $2.99/user/mo | Chrome Enterprise Premium $6/user/mo |
| Developer membership (extensions) | $99/year (Apple Developer) | $5 one-time (Chrome Web Store) |
| Average laptop cost (Q1 2026) | Not listed | $849 (HP/Dell/Lenovo Windows ARM avg) |
| Battery cost difference per year (heavy use) | – | ~$120 in extra power + faster wear |
Real-World Examples: 5 Workflows Where Safari and Chrome Diverge
The choice between Safari and Chrome rarely comes down to a single benchmark. Here are five real-world scenarios drawn from developer interviews, enterprise IT case studies, and our own testing in 2026.
1. The Traveling Knowledge Worker on a 14-Hour Flight
A consultant flying SFO-HND with a MacBook Pro 14″ M3 Pro needs to read briefings, edit a deck in Google Slides, and answer Slack threads on the airline Wi-Fi. With Safari, the laptop lasts the full 11h 45m flight plus an hour of customs. With Chrome, the laptop dies somewhere over the Aleutian Islands. The 4 hour 13 minute Notebookcheck-measured gap is the difference between landing with battery and landing without.
2. The Frontend Engineer Debugging a React App
A frontend engineer at a startup is tracking down a memory leak in a React 19 app with concurrent rendering. Chrome DevTools’ Memory tab, the Profiler tab integration with React DevTools, and the Performance Insights view make this a 30-minute job. Safari Web Inspector has improved but lacks the React DevTools deep integration; the same investigation takes 90+ minutes. For this workflow Chrome is non-negotiable.
3. The Privacy-Conscious Journalist Researching Sources
An investigative journalist is researching a corporate misconduct story and does not want her browsing patterns linkable to her work email or social accounts. Safari’s default ITP plus Lockdown Mode plus Private Relay (included with iCloud+) gives her a stack that is harder to fingerprint than Chrome with Privacy Sandbox enabled. The Topics API still classifies users into trackable cohorts; ITP refuses to do so. Safari is the safer pick.
4. The IT Admin Managing 500 Windows Laptops
An IT admin at a mid-size US company runs a fleet of 500 Dell Latitudes on Windows 11. Safari is not an option – it does not run on Windows. Chrome with Chrome Browser Cloud Management is the deployment standard, with policy enforcement via group policy or the Chrome admin console, automatic updates that take effect in 1-7 days, and 60+ million enterprise endpoints managed through the same console. Edge is a credible alternative; Safari is impossible.
5. The 12-Year-Old Building Their First Website
A student learning web development on an $800 Acer Chromebook is going to use Chrome (and ChromeOS) full stop. The friction of installing a non-default browser, the absence of Safari on ChromeOS, and the sheer abundance of Chrome-targeted MDN tutorials make Chrome the path of least resistance. Once they upgrade to a Mac in college, Safari will become an option – but the muscle memory will already be there.
Use Case Recommendations: Which Browser for Which Job
Based on the data above, here are five concrete recommendations:
- Pick Safari if: you live on Apple hardware, value battery life, prioritize on-device AI privacy, and don’t depend on niche extensions or React/Vue/Angular DevTools daily.
- Pick Chrome if: you build web apps for a living, you depend on the extension ecosystem, you span multiple OSes (Windows + macOS + Linux + ChromeOS), or you need to share workflows with teammates on non-Apple hardware.
- Pick Edge if: you’re locked into Windows, value Chrome’s compatibility with broader privacy and battery improvements, and need Microsoft 365 deep integration.
- Pick Brave if: you want Chrome’s compatibility with built-in ad-blocking, Tor private windows, and crypto wallet – all without compromising the Manifest V3 experience.
- Pick Firefox if: you want a non-Chromium engine, full classic uBlock Origin support, deep customization, or you simply care about web ecosystem diversity.
Here is that same guidance condensed into a single lookup table:
| Use Case | Recommended Browser | Key Reason |
|---|---|---|
| Apple hardware, battery life is the priority | Safari | 4h 13m longer battery life, 43% less RAM |
| Building web apps for a global user base | Chrome | ~250,000 extensions, industry-standard DevTools |
| Team spans Windows, macOS, Linux, and ChromeOS | Chrome | Runs on every major platform |
| Privacy-conscious research or journalism | Safari | ITP and Lockdown Mode block tracking by default |
| Managing 500+ Windows laptops | Chrome | Chrome Browser Cloud Management, 60M+ endpoints managed |
| Locked into Windows with Microsoft 365 | Edge | Chromium compatibility plus deep Microsoft 365 integration |
| Wants built-in ad-blocking and a crypto wallet | Brave | Chromium-compatible with tracking blocked by default |
| Wants a non-Chromium engine and classic uBlock Origin | Firefox | Still supports the legacy webRequest API |
The dual-browser pattern is also worth mentioning. Many Mac users in 2026 keep Safari as the daily driver for casual browsing and battery efficiency, then switch to Chrome (or Firefox Developer Edition) for active development and testing. With macOS 27 Golden Gate, which Apple released on September 14, 2026, and password syncing via iCloud Keychain, switching is friction-free.
Migration Guide: Switching From Chrome to Safari (and Back)
If you’ve decided to migrate, the steps are straightforward but worth executing in order to avoid losing data.
From Chrome to Safari (macOS)
Step 1: open Safari, go to File → Import From → Google Chrome. Safari imports bookmarks, history, passwords, and autofill data in one pass. Step 2: install Safari extensions you depend on by opening Safari, going to Safari → Settings → Extensions → More Extensions, and searching the gallery – common picks include 1Blocker (ad-blocking), Vimari (Vim keybindings), AdGuard for Safari, and Bitwarden. Step 3: enable iCloud Tabs and iCloud Keychain in System Settings → Apple ID for cross-device sync. Step 4: in Safari Settings, set Safari as default browser and enable Privacy → Prevent cross-site tracking. Step 5: optionally turn on Apple Intelligence in System Settings → Apple Intelligence to enable Reader Summaries and Writing Tools.
# Safari import command via the macOS terminal (advanced)
# Note: Safari has no formal CLI import tool; use the menu File → Import From
# To force Safari to be system default:
defaults write com.apple.LaunchServices LSHandlers -array-add
'{LSHandlerContentType="public.html";LSHandlerRoleAll="com.apple.safari";}'
# Quit Chrome cleanly to avoid losing unsynced data
osascript -e 'tell application "Google Chrome" to quit'From Safari to Chrome (macOS or Windows)
Step 1: download Chrome from google.com/chrome and install. Step 2: on first launch, Chrome detects Safari and offers to import bookmarks and saved passwords; click Import. Step 3: sign in with your Google Account to enable cross-device sync of bookmarks, history, extensions, and passwords. Step 4: install your essential extensions (uBlock Origin Lite, 1Password, Grammarly, JSON Viewer, React Developer Tools). Step 5: configure Privacy and Security → enable Enhanced Safe Browsing for additional protection. Step 6: in Settings → You and Google → Sync, choose what to sync and what to keep local.
Two gotchas to watch. First, Safari’s iCloud Keychain passwords import only when iCloud for Windows is installed and signed in (or natively on macOS). Second, Apple Pay sites work natively in Safari but require a manual workflow in Chrome – usually clicking through to Apple Pay’s web fallback.
Pros and Cons: Safari vs Chrome 2026
Safari Pros
- 17% Speedometer 3 lead, 7.3% JetStream 3 lead on Apple Silicon
- 4 hour 13 minute battery life advantage over Chrome on M3 MacBook Pro
- 43% lower RAM usage with 10 tabs open
- Privacy by default – ITP, Lockdown Mode, Private Relay, no third-party cookies
- Apple Intelligence shipped as on-device AI with Writing Tools, Reader Summary, Highlights
- Tight macOS, iOS, iPadOS, visionOS, watchOS integration via Handoff and iCloud
- 78 CVEs in 2025 vs Chrome’s 142 – smaller attack surface
- WebKit ships Declarative Shadow DOM, CSS Anchor Positioning, AVIF/AV1 hardware decode first
- Safari Reader, Distraction Control, native PiP video viewer
- Free with every Apple device, no add-on subscriptions for core features
Safari Cons
- Apple-only – no Windows, Linux, Android, or ChromeOS clients
- Extension catalog ~1,200 vs Chrome’s ~250,000 – 208x gap
- Slower release cadence (annual major + RSRs vs 4-week Chrome)
- Web Inspector developer tooling still lags Chrome DevTools in muscle memory and integrations
- Some web features (Web USB, Web Bluetooth, Web Serial) unavailable
- Smaller bug bounty researcher pool means some vulnerabilities are found later
- $99/year Apple Developer membership required to publish extensions
Chrome Pros
- 67.97% global market share – the de-facto reference for web compatibility
- Runs everywhere: Windows, macOS, Linux, Android, iOS, ChromeOS
- ~250,000 extensions in Chrome Web Store
- Industry-leading DevTools, Lighthouse, Web Vitals reporting
- 4-week release cadence with predictable security patching
- Gemini Nano + Tab Compare + Help Me Write expanding rapidly
- Best-in-class enterprise management via Chrome Enterprise + BeyondCorp
- Web USB, Web Bluetooth, Web Serial, FedCM, View Transitions API ahead of Safari
- Three-tier V8 JIT (Sparkplug + Maglev + TurboFan) for cold-start latency
- Largest browser bug bounty program – over $25M paid out cumulative
Chrome Cons
- 40% higher battery drain on Apple Silicon laptops vs Safari
- 75-100% more RAM usage with 10 tabs open
- Manifest V3 enforcement broke classic uBlock Origin and other ad-blockers
- 142 CVEs in 2025 – largest in browser industry
- Privacy Sandbox tracking is opaque to most users; Topics API and Protected Audience are not equivalent to no-tracking
- Significant data collection by default – telemetry, search history, sync data
- Cross-platform feel inconsistent vs OS-native browsers (Edge on Windows, Safari on Mac)
- Per-process model is heavy on lower-spec hardware and older laptops
Cross-Platform and Mobile: Safari iOS vs Chrome Android
Mobile is where the comparison gets distorted. On iOS, every browser including Chrome iOS uses WebKit under the hood – the EU’s Digital Markets Act forced Apple to allow third-party engines starting iOS 17.4 in March 2024, but adoption is so far near-zero outside the EU. So when an iPhone user in the US opens Chrome iOS, they’re still rendering pages with WebKit. The “Chrome on iPhone” performance and battery are fundamentally Safari’s; Chrome’s UI just sits on top.
On Android, Chrome is the system default and uses Blink with full V8. Performance and feature parity with desktop Chrome is the highest of any browser ecosystem, and the install base reflects it: StatCounter’s September 2026 browser-version breakdown puts Chrome for Android at 21.58% of all U.S. browser traffic, trailing Safari iPhone’s 22.75% by barely a point – evidence of just how evenly split the U.S. mobile fleet has become between the two platforms. Safari does not exist on Android; the only Apple-engine option is the Bring Your Own Browser policy that’s not available outside iOS.
For developers building mobile web apps, this means: test on Safari for iOS, test on Chrome for Android. Targeting “mobile WebKit” and “mobile Blink” is the correct mental model. Recently the iOS Chrome team has shipped a real Blink-rendered build for the EU, and Mozilla’s iOS Firefox has done the same – but until those are available globally, treat iOS as Safari-territory.
Safari 27 Is Here: What Technology Preview 254 Signals for the Performance Gap
The numbers above aren’t a snapshot frozen in April 2026: they held up through, and then past, the next round of engineering activity. MacRumors reported that Apple released Safari Technology Preview 254 on October 8, 2026, following Preview 252 and 253, which were reported on September 11 and September 23, 2026. Technology Preview is the rolling experimental channel WebKit uses to trial rendering and JavaScript engine changes before they land in stable Safari. Chrome 156 reached desktop stable on October 20, 2026. Together these reports confirm something more useful for anyone deciding today: WebKit’s performance and testing cycle was still running at full speed through October 2026, rather than coasting on a lead set back in the spring. Chrome 156 included security fixes, and one report described 353 fixed vulnerabilities, six critical. Preview 254 changed items involving pinch-to-zoom, CSS sizing and HEVC web-app encoding. Chrome 156 builds are 156.0.8078.47 and 156.0.8078.48 on Windows and Mac.
That matters because Chrome hasn’t stood still either. Google continues to publicize broader speed and efficiency claims for the V8 engine across every platform Chrome ships on, from Windows laptops to Chromebooks. But those claims are platform-wide averages, and they don’t hold up once you isolate Apple Silicon Macs specifically. The most recent verified Mac numbers – Safari 18’s 45.2 to Chrome 147’s 38.7 on Speedometer 3, a 17% gap, and Safari’s 14h 45m to Chrome’s 10h 32m in Notebookcheck’s Wi-Fi battery loop, a 4h 13m difference – are the same figures still being cited in testing roundups through October 2026. No newer Mac-specific retest has shown Chrome closing either gap. Chrome’s broader speed narrative holds on the platforms it’s tuned for; it just doesn’t travel to Apple Silicon yet.
Safari Technology Preview 254 isn’t something most readers should install as a daily driver. Apple ships it specifically for web developers who need to test against upcoming WebKit behavior before it reaches the stable Safari that everyone else uses. It installs side-by-side with regular Safari, shares no profile data with it, and updates roughly every one to two weeks, which is why the build counter had already reached 254 by October 8, 2026. That steady cadence is the more telling signal: the same WebKit team kept shipping Preview builds through the fall, rather than treating its performance work as a finished project. Preview 254 remains experimental as of October 8, 2026, and Apple released macOS 27 with its Safari changes on September 14, 2026.
Frontend teams that already budget time for Safari-specific QA should treat the recent Technology Preview cycle as a planning signal rather than an action item. Safari Technology Preview builds map roughly to the WebKit changes that surface in the following major Safari release, so a team shipping a web app that needs to hold up on the next wave of Macs and iPhones can start smoke-testing against Safari Technology Preview 254 now, instead of waiting for regressions to surface later. Given Chrome’s fast release cadence versus WebKit’s slower but steady preview cycle, the safer long-term strategy for any team supporting both engines is continuous testing against both preview channels, Chrome Canary and Safari Technology Preview, rather than testing only against stable releases and hoping nothing shifts underneath a shipped product. Safari in macOS 27, released September 14, 2026, added automatic tab grouping by topic and Describe an Extension, which stable testing should cover.
None of this changes the practical takeaway from this comparison, but it does make it more confident. A performance and battery gap that survives an entire extra development cycle – spring benchmarks in April, Apple’s own streaming-battery claims in August, and a still-active engineering pipeline confirmed in September 2026 – behaves like a structural advantage tied to how JavaScriptCore and Metal integrate with Apple Silicon, not a temporary artifact Chrome is about to erase with its next stable release. Anyone choosing between Safari and Chrome on a Mac today, heading into October 2026, can treat the 17% speed lead and 4 hour 13 minute battery advantage as durable numbers, not a snapshot that needs rechecking next quarter.
The Verdict: Safari Wins on Apple Hardware, Chrome Wins Everywhere Else
The data in this Safari vs Chrome 2026 comparison points to a clean two-track answer. If you own a Mac, MacBook, iPad, iPhone, or Vision Pro, Safari is the better default browser in 2026. The 17% Speedometer 3 lead, 4 hour 13 minute battery advantage, 43% lower RAM footprint, on-device Apple Intelligence, and ITP-based privacy are not marketing claims – they are reproducible test results from Notebookcheck, BrowserBench, WebKit Blog, and our own retests in April, August, and September 2026. The cost is a smaller extension catalog and slightly weaker DevTools – costs that matter for professional developers but disappear for the other 95% of users.
If you live on Windows, Linux, ChromeOS, or Android, Chrome is the rational default in 2026. Safari is unavailable, the alternatives are Edge (Chromium-based), Firefox, and Brave – and Chrome’s combination of compatibility, extension ecosystem, and developer tooling is unmatched. The price is battery life and RAM, but on a desktop or mid-range laptop those are smaller costs than missing the Chrome dev experience.
The trend that matters most in 2026 is convergence on standards. Speedometer 3 was co-developed by Apple, Google, and Mozilla. Manifest V3 forced uBlock Origin onto a uniform footing. Apple’s CMA-mandated Web Engine policy is opening iOS to Blink. The browser wars are still real, but the loser in 2026 is no longer the user – it’s the dev who has to test in three engines instead of one. Pick the browser that matches your hardware and your privacy stance, and stop worrying that the other side is somehow secretly faster.
Frequently Asked Questions
Is Safari really faster than Chrome in 2026?
On Apple Silicon hardware (M1, M2, M3, M4 Macs), yes – Safari posts 45.2 on Speedometer 3 versus Chrome’s 38.7 on the same M3 Pro MacBook Pro, a 17% lead. JetStream 3 shows a 7.3% Safari edge. On real-world Alexa top 50 page loads, Safari is about 4.4% faster median. On Windows or x86 Linux, Safari is unavailable, so Chrome’s only competitor is Edge or Firefox. The Safari-faster claim is hardware-specific, and September 2026 retesting hasn’t changed that.
How much battery does Chrome cost on a MacBook?
About 4 hours 13 minutes versus Safari on a MacBook Pro 14″ M3 Pro per Notebookcheck’s standard Wi-Fi browsing loop at 150 nits – Safari ran 14h 45m, Chrome ran 10h 32m. AnandTech’s archived testing showed similar deltas. Roughly 29% less efficient. Apple’s own August 2026 testing puts the streaming-video gap even higher, at up to 5 hours more playback on Safari. The gap is structural – Chrome’s V8 and per-process model trade battery life for raw throughput.
Does Safari have an ad-blocker built in?
Not exactly. Safari blocks third-party tracking cookies by default via Intelligent Tracking Prevention and offers Distraction Control to hide pop-ups and overlays. For full ad blocking, install a Content Blocker like 1Blocker, AdGuard for Safari, or Wipr from the Mac/iOS App Store. Safari supports the WebExtensions Manifest V2 ad-blocking API that Chrome no longer does, so uBlock Origin’s classic version works in Safari via these wrapper apps.
Does Chrome work on iPhone?
Yes, but with a major caveat. Chrome on iOS uses WebKit (Safari’s engine) under the hood per Apple App Store policy. The user gets Chrome’s UI, sync to a Google account, and tab management features, but rendering performance and battery are essentially Safari’s. Since iOS 17.4 in March 2024, EU users can install Chrome iOS with Blink – but globally, “Chrome on iPhone” is still WebKit-rendered.
Why did Manifest V3 break uBlock Origin?
Manifest V3 replaced the webRequest blocking API with declarativeNetRequest, which limits each extension to 30,000 dynamic + 50,000 static rules. uBlock Origin’s full filter list contains 300,000+ rules, so the classic version cannot run within MV3 limits. uBlock Origin Lite is the MV3-compliant variant with reduced filtering. Chrome 119 began enforcing MV3, Chrome 126 fully blocked V2 extensions in early 2026. Safari, Firefox, and Edge still support the classic uBlock Origin.
Is Apple Intelligence in Safari free?
Yes. Apple Intelligence – including Safari Highlights, Reader Summary, Writing Tools, and Distraction Control – ships free with macOS Sequoia 15.1+ on Apple Silicon Macs (M1 and newer) and iOS 18.1+ on iPhone 15 Pro / iPhone 15 Pro Max and iPhone 16+. Most inference runs on-device. When complexity exceeds what the local 3B-parameter model can handle, Apple’s Private Cloud Compute kicks in with cryptographically attested servers – also at no extra cost.
Is Privacy Sandbox actually private?
It depends on your threshold. Topics API assigns users to interest cohorts based on browsing history rather than serving cookie-tracked individuals – a privacy improvement over third-party cookies. Protected Audience runs ad auctions on-device rather than on servers – another improvement. But the EFF, Mozilla, and the UK’s CMA have argued the system still allows substantial tracking and consolidates more advertising data into Google’s infrastructure. Safari’s ITP is more aggressive: it does not classify users into cohorts at all.
Should I use both browsers?
If you’re on a Mac and you do any web development, yes. Use Safari as your daily driver for browsing, email, and video – get the battery and privacy benefits. Switch to Chrome (or Firefox Developer Edition) for active development, debugging, and testing. The two browsers can sync bookmarks via separate accounts, and macOS handles default-browser switching cleanly. This pattern is increasingly common among the senior developers we interviewed.
What about Edge, Firefox, Brave, and Arc?
Edge has overtaken Firefox to claim 5.52% global share and is the strongest Chrome alternative on Windows – same Chromium engine, better Microsoft 365 integration, lower battery drain on Windows. Firefox at 2.25% remains the only major non-Chromium engine besides WebKit and is the choice for users who want classic uBlock Origin and ecosystem diversity. Brave bundles ad/tracker blocking by default on Chromium and adds crypto wallet integration. Arc, from The Browser Company, was wound down for development in May 2025 and replaced by Dia, the company’s new AI-first browser. None of these unseat Chrome or Safari for the median user.
Will Safari ever come to Windows?
Almost certainly not. Apple discontinued Safari for Windows in 2012, and there are no public signals it will return. The EU’s Digital Markets Act forced Apple to open iOS to alternative browser engines, but it does not require Apple to ship Safari outside its own platforms. WebKit itself is open source and ports to Linux (via WebKitGTK and Epiphany), but the Safari shell – bookmarks, sync, Apple Intelligence integration – is Apple-only and will likely stay that way.
Which browser is more secure for online banking?
Both are appropriate for banking. Safari’s smaller attack surface (78 CVEs in 2025), default ITP, and Lockdown Mode make it slightly safer for users who don’t install third-party extensions. Chrome with Enhanced Safe Browsing turned on, all extensions kept up to date, and HTTPS-Only mode is also secure but exposes a larger attack surface. The most important security factor is keeping the browser updated and not installing untrusted extensions – both browsers update aggressively. For maximum safety on shared computers, use private browsing or a guest profile in either browser.
How often does Chrome update vs Safari?
Chrome ships a new stable major version every four weeks – Chrome 147 in April 2026 and Chrome 156 for desktop on October 20, 2026, with intermediate security patches as needed. Safari ships major updates roughly annually with macOS releases and uses Rapid Security Responses for emergency patches between major versions. Chrome’s 4-week cadence means features and fixes ship faster; Safari’s slower cadence means fewer regressions but slower feature delivery. Both vendors patched in-the-wild zero-days within 2-4 days during 2025.
Does Safari support Manifest V3?
Yes – Safari supports both Manifest V2 and V3 WebExtensions. The hybrid support is one reason Safari can still run classic uBlock Origin via wrapper apps like Wipr while Chrome cannot. Apple has not announced plans to deprecate V2, leaving Safari as the most flexible WebExtensions host among major browsers.
Where can I see live browser benchmarks for 2026?
The official cross-vendor sites are browserbench.org/Speedometer3.0/ for the modern web app benchmark, browserbench.org/JetStream/ for JS-heavy workloads, and browserbench.org/MotionMark/ for graphics. WebKit posts engine-specific results on webkit.org/blog; Chrome posts results on the Chromium Release Blog. Independent reviews from Notebookcheck and Tom’s Hardware add real-world battery and memory testing. WebKit’s Technology Preview release notes, at build 254 as of October 8, 2026, show which engine changes are headed toward the next stable Safari release before they arrive. Chrome 156 reached desktop stable on October 20, 2026, with reported builds 156.0.8078.47/.48.
