Citation and evidence

Gemini 3.5 Flash

17 min full readUpdated 28 references

This article's verification

Report a problem with this article

Edit article
More

Use this article

Raw MarkdownExplore connections

Improve this page

Revision historyDiscuss this article

Topics

Google DeepMindLarge Language ModelsMultimodal AI

Cite this article

Gemini 3.5 Flash is a fast frontier large language model developed by Google DeepMind, announced at Google I/O 2026 on May 19, 2026 and made generally available the same day [1][2]. It is the first model in the Gemini 3.5 family and the successor to Gemini 3 Flash, which shipped in December 2025 [3][4]. Google positioned the model around agentic execution and coding rather than conversational chat, describing it as the company's "most intelligent Flash model" and claiming it surpasses the larger Gemini 3.1 Pro on most coding and agentic evaluations and produces output tokens about four times faster than other frontier models [1][5]. Google deprecated the model on October 8, 2026, routing its API requests to Gemini 3.6 Flash [27][28].

API prices and benchmark results

Compare all listed models

Standard text API rates in USD per 1 million tokens. Named benchmark results are not a universal ranking.

List prices: AI Compute Tracker, owned and operated by the same team as AI Wiki. Routed rates: OpenRouter.

This panel uses the same automatically refreshed snapshot as the front page. Benchmark versions or test configurations may differ or be unknown; scores are not ranked.
ModelInput / 1MOutput / 1MBenchmark resultExample cost
Gemini 3.5 FlashgoogleOpenRouter API rate$1.5Cached: $0.15$9Not measured in this snapshotExample cost$7.5
Sources & rate details for Gemini 3.5 Flash
OpenRouter catalogueSource fetched UTC

OpenRouter listed base text-token routing rates in USD per 1M tokens, not a provider-direct quote. Provider routing, context tiers, caching, reasoning, tools, media, and other fees can change the actual bill. Batch/free/router variants are excluded.

Example cost: 1,000 requests, each with 2,000 input and 500 output tokens, at the listed standard text rates. No caching, batch discount, long-context surcharge, tools, media or tax; not a capability or availability check. Adjust this workload in the cost planner (choose a model there).

Compare and deploy Gemini 3.5 Flash

These sourced facts are shared with the comparison and planning tools. Each field retains its own review date.

Gemini 3.5 Flash

Article updated after the specification review. Inspect the sources before relying on these fields; a newer edit does not necessarily change the model facts.

Provider API lifecycle
Deprecated; routed to Gemini 3.6 Flash since Oct 8, 2026Source · checked 2026-10-15 · review due
Context tokens
1.04MSource · checked 2026-07-23

Deprecated October 8, 2026; requests now go to Gemini 3.6 Flash, and Google's pricing page no longer lists 3.5 Flash.

Gemini family logo: gradient four-point star beside the Gemini wordmark
Gemini family logo, not model-specific artwork. Source: Google. Gemini is a trademark of Google LLC. AI Wiki is not endorsed by or affiliated with Google.[24] Image: Google. Google Image Library publication permission.

Overview

Gemini 3.5 Flash sits in the "Flash" tier of the Gemini line, the segment Google tunes for low latency and high throughput rather than maximum raw capability. With this release Google argued that the speed-optimized tier had caught up to, and on several axes overtaken, the previous flagship Pro model [1][6]. The pitch centered on long-horizon agentic work: tasks where a model plans, calls tools, writes and runs code, and iterates over many steps with limited human input [5].

The model is multimodal on input. It accepts text, images, video, audio, and PDF documents, and returns text [7]. It carries a one-million-token context window and supports tool use including Google Search grounding, code execution, file search, URL context, and function calling [8]. At launch Google's documentation listed no computer-use support for Gemini 3.5 Flash, although the current version of the same documentation lists Computer Use among its supported tools [8][12].

Announcement and release

Google unveiled Gemini 3.5 Flash during the Google I/O 2026 developer keynote on May 19, 2026, and shipped it to production the same day rather than as a staged preview [1][2]. TechCrunch described the launch as Google's shift from pitching AI as a conversational tool to an agentic one, and Koray Kavukcuoglu, DeepMind's chief technologist, said the model offers "an incredible combination of quality and low latency" [2]. TechCrunch reported that the model can run autonomously for multiple hours, and Tulsee Doshi, Google's senior director and head of product, said it will at times pause to ask for user input at decision points [2].

At launch the model carried the API identifier gemini-3.5-flash and replaced the earlier preview identifier used for Gemini 3 Flash [8]. Google described it as generally available, stable, and ready for scaled production use [9]. The same release powered consumer-facing features announced at I/O, including the model behind the Gemini app and AI Mode in Google Search [2].

Place in the Gemini line

Gemini 3.5 Flash is the opening model of the Gemini 3.5 generation. Its direct predecessor in the Flash tier is Gemini 3 Flash (released December 17, 2025), and the wider 3.x family includes Gemini 3 Pro and the intermediate Gemini 3.1 Pro that Google used as its comparison baseline at launch [3][4][6]. Google reported that 3.5 Flash outperforms Gemini 3.1 Pro on challenging coding and agentic benchmarks, which some outlets described as the first time a Flash-tier model had surpassed a Pro-tier model on those workloads [3].

A larger sibling, Gemini 3.5 Pro, was previewed as "rolling out next month," with reporting putting that timeline in June 2026 [1][9][10]. It did not ship on that schedule: Google's July 21 post said 3.5 Pro was "currently testing with partners," and by November 2026 the Gemini API's model list led with Gemini 4 Argon rather than a 3.5 Pro [14][26]. Some coverage noted audience disappointment that the Pro variant was not shipping at the event itself [10]. For background on the broader generation, see Gemini 3.

Architecture and training

Google disclosed little about the underlying architecture or training data. The published model card and developer documentation describe the model's interface and behavior rather than its internals, and Google did not release parameter counts or training-corpus details [7][8]. The stated knowledge cutoff is January 2025 [8][12].

What Google did detail is the reasoning interface. Gemini 3.5 Flash uses a thinking_level parameter with values minimal, low, medium, and high, where medium is the default. This changed the default from the high setting used by Gemini 3 Flash, and Google said the low setting was retuned to be markedly stronger on code and agentic tasks [8]. The model also preserves intermediate reasoning across turns of a multi-turn conversation automatically, a behavior Google labels "thought preservation" [8].

Capabilities and modalities

The model accepts text, image, video, audio, and PDF input and produces text output [7]. Google lists its primary strengths as everyday tasks, agentic coding, advanced reasoning, multimodal understanding, and long-context understanding [7]. Tool capabilities include function calling, structured output, Google Search and Google Maps grounding, file search, URL context, and code execution [7][8].

The agentic framing was the centerpiece. Google described the model as able to execute long, multi-step workflows, and reporting cited early uptake among banks and fintechs automating multi-week processes [2][5]. On raw output speed, Kavukcuoglu said the model is four times faster than other frontier models and that Google had "taken it to another level" with an optimized version that is 12 times faster with the same quality [2].

Benchmark performance

Google reported that Gemini 3.5 Flash leads Gemini 3.1 Pro across most coding, agentic, and multimodal evaluations, though it trails on a few reasoning and long-context tests. The table below lists scores published by Google and compiled in independent launch analyses; comparison figures are versus Gemini 3.1 Pro [1][7][6].

BenchmarkCategoryGemini 3.5 FlashGemini 3.1 Pro
Terminal-Bench 2.1Agentic coding76.2%70.3%
SWE-Bench Pro (public)Coding55.1%54.2%
MCP AtlasMulti-step tool use83.6%78.2%
ToolathlonReal-world tool use56.5%Not reported
OSWorld-VerifiedComputer tasks78.4%76.2%
Finance Agent v2Agentic finance57.9%43.0%
GDPval-AAAgentic (Elo)16561314
CharXiv ReasoningChart reasoning84.2%83.3%
MMMU-ProMultimodal reasoning83.6%80.5%
Blueprint-Bench 2Multimodal33.6%26.5%
MRCR v2 (128k)Long context77.3%84.9%
MRCR v2 (1M)Long context26.6%26.3%
Humanity's Last ExamReasoning40.2%44.4%
ARC-AGI-2Reasoning72.1%77.1%

Expanded article table

At launch, the high-effort configuration of Gemini 3.5 Flash scored 55 on the Artificial Analysis Intelligence Index, and Artificial Analysis measured over 280 output tokens per second in pre-release testing; its live model page later showed about 168 tokens per second [25][11]. Google highlighted four headline numbers from the model card: 76.2% on Terminal-Bench 2.1, 83.6% on MCP Atlas, 84.2% on CharXiv Reasoning, and an Elo of 1656 on GDPval-AA [1][7].

Availability, context window, and pricing

Gemini 3.5 Flash launched across Google's developer and enterprise surfaces as well as its consumer apps. On October 8, 2026, Google deprecated gemini-3.5-flash: it is no longer selectable in Google AI Studio, and the endpoint stays up but automatically routes API requests to Gemini 3.6 Flash [27][28]. At launch it was available through the Google AI Studio interface and the Gemini API, in Android Studio, on Google Vertex AI and the Gemini Enterprise Agent Platform, in Google Antigravity, and to everyone in the Gemini app and AI Mode in Search [1][2][7].

The context window is 1,048,576 input tokens with a maximum of 65,536 output tokens [8][12]. At launch, pricing was volume-based per million tokens; rates differed slightly between Google's global and non-global serving regions, and batch processing was offered at a discount [4][12][13]. Google's pricing page no longer lists the model after its deprecation [16].

ItemRate (per 1M tokens)
Input (global)$1.50
Output (global)$9.00
Cached input$0.15
Cache storage$1.00 per 1M token-hours
Input / output (non-global)$1.65 / $9.90
Batch input / output$0.75 / $4.50

Expanded article table

These rates are higher than the prior generation. Simon Willison noted the model is about three times the price of Gemini 3 Flash Preview and six times that of Gemini 3.1 Flash-Lite, and read the increase as the major labs probing how much their API customers will pay [12]. Willison found it notable that, despite the higher per-token cost, Google rolled the model out across so many of its free consumer products [12].

Reception

Coverage of the launch split between the benchmark story and the pricing story. Outlets including TechCrunch and MarkTechPost emphasized Google's claim that a Flash-tier model now beat its own Pro-tier model on coding and agentic work, and the explicit reorientation toward agents over chatbots [2][5]. The model landed in the top-right quadrant of Artificial Analysis's intelligence-versus-speed chart, the region Google wanted to occupy [1][11].

Other commentary was cooler. Trending Topics characterized the release as "more of a solid incremental improvement than a milestone" and pointed out that the version number is 3.5, not 4, while flagging the "noticeable" price increase over the predecessor [10]. The same report noted a muted early standing on community arena rankings and described audience disappointment that Gemini 3.5 Pro was delayed rather than shipped at the event [10].

July 2026 variants

Google expanded the Gemini 3.5 family on July 21, 2026 with Gemini 3.5 Flash-Lite and Gemini 3.5 Flash Cyber. Despite the shared version number, the two releases have different relationships to Gemini 3.5 Flash. Flash-Lite is a generally available efficiency model based on Gemini 3.1 Flash-Lite, while Flash Cyber is a specialized fine-tune of Gemini 3.5 Flash that Google announced for a forthcoming restricted pilot [14][17][20]. Independent launch coverage likewise treated Flash-Lite as a public high-volume model and Flash Cyber as a security model limited to selected partners [23].

ModelRelationshipPublic identifierStatus on July 24, 2026Primary role
Gemini 3.5 FlashBase modelgemini-3.5-flashGenerally availableAgentic execution, coding, and multimodal work
Gemini 3.5 Flash-LiteBased on Gemini 3.1 Flash-Litegemini-3.5-flash-liteGenerally availableHigh-volume, latency-sensitive, and low-cost work
Gemini 3.5 Flash CyberFine-tuned from Gemini 3.5 FlashNo public standalone identifierForthcoming limited CodeMender pilotVulnerability discovery, validation, and patching

Expanded article table

Gemini 3.5 Flash-Lite

Gemini 3.5 Flash-Lite was released as a stable production model on July 21. Google Cloud lists it as generally available and gives a retirement date of July 21, 2027 or later [18]. It accepts text, images, video, audio, and PDFs, produces text, and has the same published token limits as the base model: 1,048,576 input tokens and 65,536 output tokens [15]. Its intended workloads include translation, classification, document processing, agentic search, and subagent execution [14][17].

The model supports thinking, caching, structured output, code execution, file search, function calling, URL context, and grounding with Google Search and Google Maps. It does not support audio generation, image generation, or the Live API [15]. As of November 2026, both the Gemini Developer API page and the Gemini Enterprise Agent Platform page list Computer Use as supported, and Google's launch post described computer use as a built-in tool for the model [14][15][18]. Google set the default thinking level to minimal for latency and cost. Its Cloud guidance recommends medium or high for autonomous subagents because minimal thinking can end multi-step tool work prematurely [18].

Standard paid Gemini API pricing is $0.30 per million input tokens and $2.50 per million output tokens, including thinking tokens. A standard cache hit costs $0.03 per million tokens, with storage billed at $1.00 per million token-hours. Batch and Flex processing each cost $0.15 per million input tokens and $1.25 per million output tokens; Priority processing costs $0.54 and $4.50 respectively [16]. These rates place Flash-Lite below the base model's $1.50 input and $9.00 output rates, although the default reasoning settings and target workloads also differ.

Flash-Lite evaluation and safety

Google evaluated Flash-Lite across coding, agentic, multimodal, and long-context tests. The published results below compare it with Gemini 3.1 Flash-Lite, the model on which it is based [17]. They are vendor-reported scores, and results can depend on the named harness, tool access, and context length.

Benchmark and configurationGemini 3.5 Flash-LiteGemini 3.1 Flash-Lite
SWE-Bench Pro (Public)54.2%38.3%
Terminal-bench 2.1, Terminus-2 harness54.0%31.0%
MLE-Bench39.2%22.0%
GDPVal-AA v21140 Elo642 Elo
OSWorld-Verified74.0%54.3%
CharXiv Reasoning, no tools74.5%73.2%
CharXiv Reasoning, with tools76.5%75.6%
GDM-MRCR v2, 128k average72.2%60.1%
GDM-MRCR v2, 1M pointwise21.3%12.3%

Expanded article table

Google's launch post cited an Artificial Analysis speed measurement of 350 output tokens per second [14]. As of early November 2026, the Artificial Analysis page reported 317 output tokens per second and a 5.11-second time to first answer token under its own test configuration [19]. The measurements were recorded under different configurations or at different times, so neither figure should be read as a fixed throughput guarantee.

The Flash-Lite model card lists hallucinations and occasional slowness or timeouts among the known limitations, with a March 2026 knowledge cutoff. Google's automated internal evaluation found better text-to-text and multilingual safety than Gemini 3.1 Flash-Lite, no change in image-to-text safety, improved refusal tone, and a 5.32-percentage-point regression in unjustified refusals, where a lower score is better. Google also reported specialist red teaming outside the model development team, satisfaction of child-safety launch thresholds, and no egregious concerns [17]. Its frontier-safety conclusion was inferred from Gemini 3.1 Pro's assessment rather than derived from a separately published full threshold table for Flash-Lite.

Gemini 3.5 Flash Cyber

Gemini 3.5 Flash Cyber is a cybersecurity-focused variant built on top of Gemini 3.5 Flash and fine-tuned to find, validate, and patch software vulnerabilities [14][20]. It runs inside CodeMender, where multiple subagents can inspect code paths and combine their findings into one report. Google reports using the system internally across Chrome, Android, Cloud, Ads, and YouTube codebases [20].

The announcement was not a public API release. Google said Flash Cyber would become available "soon" in a limited-access CodeMender pilot restricted to governments and trusted partners, with access expanding over time. The company cited the technology's dual-use risk as the reason for the controlled deployment [14][20]. General CodeMender capabilities offered through the Gemini Enterprise Agent Platform with generally available Gemini models are separate from the Flash Cyber pilot [20][22].

Google has not published a standalone Flash Cyber API identifier, numerical token price, context or output limits, modality specification, general-release date, or dedicated model card [20][21][22]. Although the launch post says it has a lower token price than larger cybersecurity models, that statement does not establish a public rate. Specifications and safety findings from the base Gemini 3.5 Flash therefore cannot be assumed to apply unchanged to the specialized model.

Cyber evaluation and restrictions

Google's headline CyberGym result measures the combined CodeMender system, not one raw model call. CodeMender could invoke Flash Cyber up to five times before producing a final report, and the system reached an 83.2% pass@1 success rate on a benchmark covering hundreds of real-world software vulnerabilities. Google noted that competitor figures in the comparison chart were self-reported by their providers [20].

EvaluationConfigurationReported result
CyberGymCodeMender with up to five Flash Cyber calls per final report83.2% pass@1
V8 JavaScript engineFixed number of invocations55 unique confirmed issues, versus 47 for Gemini 3.5 Flash and 36 for Claude Opus 4.6
V8 unique findingsSame fixed-invocation test10 issues not found by either comparison model
Big Sleep evaluationComplex Chrome and Safari code without safety guardrailsGoogle reported a substantial lead over mainline Gemini 3.5 Flash and Gemini 3.6 Flash, without publishing a numerical score in the article text

Expanded article table

The Big Sleep test was built by another Google team and intentionally removed safety guardrails, so it was an internal stress test rather than an independent evaluation of deployed pilot behavior [20]. Google also used private, undisclosed vulnerabilities in Chrome commit scanning to reduce contamination, but it did not publish enough dataset detail for outside reproduction. TechRepublic noted further evidence gaps, including unpublished false-positive, patch-acceptance, and regression rates and the absence of independent production testing [22]. The limited access policy reduces exposure to misuse, but no dedicated public frontier-safety assessment for the specialized fine-tune accompanied the announcement.

Limitations

Several limits were disclosed or evident at launch. Computer Use was not supported at launch, unlike some earlier Gemini models, although Google's documentation later listed it as supported [8][12]. The model output is text only, so it cannot generate images or audio directly [7]. On a handful of evaluations it trails Gemini 3.1 Pro, including Humanity's Last Exam (40.2% versus 44.4%), ARC-AGI-2 (72.1% versus 77.1%), and the 128k-token slice of MRCR v2 (77.3% versus 84.9%), which suggests the gains concentrated in agentic and coding workloads rather than across the board [6]. As with prior Gemini releases, Google did not publish architecture or training-data details, so independent verification of those aspects is not possible [7][8].

Continue reading

More to read

Topics:Google DeepMindLarge Language ModelsMultimodal AI

References

  1. ^1 ^2 ^3 ^4 ^5 ^6 ^7 ^8 ^9Google DeepMind. "Gemini 3.5: frontier intelligence with action." blog.google, May 19, 2026. blog.google/...gemini-3-5
  2. ^1 ^2 ^3 ^4 ^5 ^6 ^7 ^8 ^9Maxwell Zeff. "With Gemini 3.5 Flash, Google bets its next AI wave on agents, not chatbots." TechCrunch, May 19, 2026. techcrunch.com/...t-ai-wave-on-agents-not-chatbots
  3. ^1 ^2 ^3DataNorth AI. "Google Releases Gemini 3.5 Flash: Frontier-Level Coding and Agentic Performance at 4x Speed." datanorth.ai/...google-releases-gemini-3-5-flash
  4. ^1 ^2 ^3Digital Applied. "Gemini 3.5 Flash: Benchmarks, Thinking & API Guide 2026." digitalapplied.com/...5-flash-benchmarks-api-guide
  5. ^1 ^2 ^3 ^4MarkTechPost. "Google Introduces Gemini 3.5 Flash at I/O 2026: A Faster and Cheaper Model for AI Agents and Coding." May 20, 2026. marktechpost.com/...model-for-ai-agents-and-coding
  6. ^1 ^2 ^3 ^4LLM Stats. "Gemini 3.5 Flash: Benchmarks, Pricing, and Complete Specs." llm-stats.com/...gemini-3.5-flash-launch
  7. ^1 ^2 ^3 ^4 ^5 ^6 ^7 ^8 ^9 ^10Google DeepMind. "Gemini 3.5 Flash - Model Card." May 19, 2026. deepmind.google/...gemini-3-5-flash
  8. ^1 ^2 ^3 ^4 ^5 ^6 ^7 ^8 ^9 ^10 ^11Google. "What's new in Gemini 3.5 Flash." Gemini API documentation. ai.google.dev/...whats-new-gemini-3.5
  9. ^1 ^2Google. "Gemini 3 Developer Guide." Gemini API documentation. ai.google.dev/...gemini-3
  10. ^1 ^2 ^3 ^4Trending Topics. "Google Launches Gemini 3.5 Flash With Higher Prices but No Generational Leap." trendingtopics.eu/...ices-but-no-generational-leap
  11. ^1 ^2Artificial Analysis. "Gemini 3.5 Flash: Intelligence, Performance & Price Analysis." artificialanalysis.ai/...gemini-3-5-flash
  12. ^1 ^2 ^3 ^4 ^5 ^6 ^7Simon Willison. "Gemini 3.5 Flash: more expensive, but Google plan to use it for everything." May 19, 2026. simonwillison.net/...gemini-35-flash
  13. ^OpenRouter. "Gemini 3.5 Flash - API Pricing & Benchmarks." openrouter.ai/...gemini-3.5-flash
  14. ^1 ^2 ^3 ^4 ^5 ^6 ^7Google. "Introducing Gemini 3.6 Flash, 3.5 Flash-Lite, and 3.5 Flash Cyber." July 21, 2026. blog.google/...lash-3-5-flash-lite-3-5-flash-cyber
  15. ^1 ^2 ^3Google. "Gemini 3.5 Flash-Lite." Gemini API model documentation. ai.google.dev/...gemini-3.5-flash-lite
  16. ^1 ^2Google. "Gemini Developer API pricing." ai.google.dev/...pricing
  17. ^1 ^2 ^3 ^4Google DeepMind. "Gemini 3.5 Flash-Lite" (model card). Published July 21, 2026. deepmind.google/...gemini-3-5-flash-lite
  18. ^1 ^2 ^3Google Cloud. "Gemini 3.5 Flash-Lite." Gemini Enterprise Agent Platform documentation. docs.cloud.google.com/...3-5-flash-lite
  19. ^Artificial Analysis. "Gemini 3.5 Flash-Lite Intelligence, Performance & Price Analysis." artificialanalysis.ai/...gemini-3-5-flash-lite
  20. ^1 ^2 ^3 ^4 ^5 ^6 ^7 ^8Google DeepMind. "Introducing Gemini 3.5 Flash Cyber." July 21, 2026. deepmind.google/...roducing-gemini-3-5-flash-cyber
  21. ^Google DeepMind. "Gemini 3.5 Flash Cyber." deepmind.google/...cyber
  22. ^1 ^2 ^3TechRepublic. "Google Holds Back Gemini 3.5 Flash Cyber as CodeMender Enters Preview." July 23, 2026. techrepublic.com/...-google-flash-cyber-codemender
  23. ^Axios. "Google releases series of new cheaper Gemini models." July 21, 2026. axios.com/...google-gemini-ai-models
  24. ^Google. "Gemini logo (Google Image Library)". Google Image Library publication permission. Original official raster, unchanged.
  25. ^Artificial Analysis. "Gemini 3.5 Flash: The new leader in intelligence versus speed." May 19, 2026. artificialanalysis.ai/...erything-you-need-to-know
  26. ^Google. "Models." Gemini API documentation. ai.google.dev/...models
  27. ^1 ^2Google. "Release notes" (October 8, 2026 entry). Gemini API changelog. ai.google.dev/...changelog
  28. ^1 ^2Google. "Gemini deprecations." Gemini API documentation. ai.google.dev/...deprecations

Improve this article

Add missing citations, update stale details, or write a clearer explanation. Anyone can edit: changes go live when you publish, and every edit is checked by an editor.

6 revisions · v6 · 3,368 words · full history

Fact-checks are independent of edits: a reviewer re-verifies the article against its sources and stamps the date. How we verify

Research and drafting on this wiki are AI-assisted, under named human editorial standards. How AI is used here

Reviewer note: Independent fact-check of the full article against its cited sources (110 claims, 31 sources fetched); 5 material + 13 minor fixed

Cite this page: AI Wiki. "Gemini 3.5 Flash." aiwiki.ai, updated 24 Jul 2026, fact-checked 4 Nov 2026. CC BY 4.0. https://aiwiki.ai/wiki/gemini_3_5_flash

Edit article