Gemini 3.8 Flash: Pricing, Specs, 3 Regressions

Google shipped Gemini 3.8 Flash on 2 September 2026, three weeks after Gemini 3.7 Flash. The unusual part is the price tag: there isn't a new one. Every tier, every modality, the same context window, the same output ceiling. On paper, 3.8 Flash looks like a free upgrade.

It isn't. The list price is identical, but the effective price is not — 3.8 Flash reasons harder per task, and Artificial Analysis now measures it at $1.24 per Intelligence Index task against 3.7 Flash's $0.93. It is also slower on both latency axes. This guide separates what genuinely improved from what the unchanged price tag hides, keeps Google's own claims apart from third-party measurements, and stamps the benchmark version on every independent figure — because Artificial Analysis rebased its Intelligence Index in October 2026 and the widely-quoted older numbers do not convert.

What is Gemini 3.8 Flash?

Gemini 3.8 Flash is the current stable model in Google's Flash line — the tier positioned between the cheap Flash-Lite models and the frontier Pro/Ultra models. Google's Gemini API release notes describe the 2 September GA release as "our most intelligent Flash model, engineered for long-horizon software engineering, autonomous agents, and complex enterprise workflows."

It is a reasoning model by default. Thinking cannot be switched off entirely; it can only be dialled between three levels.

Two absences are worth flagging before you plan around this model. There is no Live API support, so 3.8 Flash cannot drive a bidirectional realtime voice session — it ingests audio as a file, it does not converse over a stream. And it produces text only: no image generation, no speech output. Those live on Google's separate Omni and image models.

Google also does not publish a knowledge cutoff date. The official model page lists every token limit and capability flag but omits the cutoff, as it does for 3.7 Flash. If your workload depends on recent world knowledge, use search grounding or URL context rather than assuming a cutoff you can't verify.

What changed from Gemini 3.7 Flash?

Start with what didn't change, because it's almost everything. Gemini 3.7 Flash, released 13 August 2026, has the identical model ID pattern, the identical 1,048,576-token context, the identical 65,536-token output ceiling, the identical five input modalities, the identical capability matrix, and — verified against Google's pricing page — the identical price on every single tier.

So the delta is entirely in model quality, and it is real but uneven. Here is the honest picture:

The pattern is clear once you line it up: 3.8 Flash buys quality with latency and tokens. Every quality measure moves the right way — two points on the independent Intelligence Index, and clean wins on all three benchmarks Google chose to publish. Human preference is directionally up but not settled: the LMArena gap is seven points with ±5 bars on both sides, so the intervals still overlap, and both scores remain marked preliminary.

Against that sit three regressions, and Google mentions none of them.

• Output speed fell from 279.7 to 239.1 tokens per second. 3.8 Flash streams roughly 15% slower than the model it replaces — the opposite of what a Flash-tier refresh is supposed to do.
• Time to first token got worse, from 10.62 to 13.45 seconds. That is 2.8 extra seconds before anything appears, a 27% increase, and the direct consequence of reasoning harder by default.
• Cost per completed task rose 33%, from $0.93 to $1.24 across Artificial Analysis's evaluation suite — even though the per-token price never moved. Thinking tokens are billed at the output rate, so a model that deliberates more costs more at an identical rate card.

That last one is the trap. "Same price" is true per token and false per task, and per task is what lands on your invoice.

On the Design Arena website category, OpenRouter's mirrored scores put 3.8 Flash at 1311 Elo and a 56.3% win rate against 3.7 Flash's 1318 Elo and 57.1%. Slightly behind its predecessor on front-end design taste.

Google's most eye-catching claim is a customer quote, not a benchmark. On the DeepMind Flash page, Glean's Thai Tran says 3.8 Flash completed "more than three times as many tasks as Gemini 3.7 Flash in our evaluations." That is a document-heavy workload at one company on an undisclosed harness — directionally consistent with the agentic index jump, but not a number you should expect to reproduce.

We cover the head-to-head in depth in Gemini 3.8 Flash vs 3.7 Flash. For the previous generation's launch context, see our Gemini 3.7 Flash launch guide.

What does Gemini 3.8 Flash cost, really?

The list prices are simple. What people get wrong is the effective price, because of how thinking tokens are billed.

Two things in that table are easy to miss.

First, Google's pricing page states the output price "includes thinking tokens." Gemini 3.8 Flash reasons by default at medium, and every one of those internal tokens is billed at $3.75 per million — the output rate, not the input rate. You never see them in the response body, but you pay for them. The thinking documentation is explicit: "response pricing is the sum of output tokens and thinking tokens," and exposes the count as total_thought_tokens.

Second, the current prices are promotional through 31 December 2026. Google's pricing page shows standard rates doubling on 1 January 2027 — to $1.50 input and $7.50 output. If you are modelling annual spend, model two halves of the year.

A worked example that includes the reasoning tokens

Take a realistic agentic call: 30,000 input tokens of context, 800 tokens of visible answer, and 6,000 thinking tokens at the default medium level.

• Input: 30,000 × $0.75 / 1M = $0.0225
• Output + thinking: (800 + 6,000) × $3.75 / 1M = $0.0255
• Total: $0.0480 per call

The naive estimate — the one almost every cost calculator produces — counts only the 800 visible output tokens and lands on $0.0255. The real bill is 1.9× that. At 1,000 calls a day, the difference is roughly $22 a day, or about $660 a month, on a workload you had budgeted at half the price.

Three levers actually move this number:

• Drop thinking_level to low on tasks that don't need deliberation. This directly cuts the most expensive token class. It is the single highest-leverage cost control on this model.
• Use the Batch API for anything not user-facing. Half price on both input and output takes the example call from $0.0480 to $0.0240.
• Cache the stable prefix. Cached input reads cost $0.075 per 1M — a 10× discount. If 25,000 of the 30,000 input tokens are a fixed system prompt and document set, input drops from $0.0225 to $0.0056, and the call lands near $0.031. Watch the $0.50 per 1M-tokens-per-hour storage fee: caching only pays off if you re-read the cache often enough to beat it.

How good is Gemini 3.8 Flash?

Google's own numbers, from the DeepMind Flash page, position 3.8 Flash at the top of a field that includes frontier models several times its price:

• HLE-Verified: 54.9% — ahead of GPT-5.6 Sol (54.5%), Claude Opus 5 (54.4%), Gemini 3.7 Flash (53.6%), GPT-5.6 Terra (51.1%) and Claude Sonnet 5 (31.0%).
• Vals Finance Agent v2: 61.4% — ahead of 3.7 Flash (59.0%), Claude Opus 5 (58.6%), GPT-5.6 Terra (54.4%), Claude Sonnet 5 (53.9%) and GPT-5.6 Sol (53.8%).
• Harvey Legal Agent: 10.0% — ahead of 3.7 Flash (8.8%), Claude Opus 5 (6.7%), Claude Sonnet 5 (5.0%), GPT-5.6 Sol (2.5%) and GPT-5.6 Terra (0.8%). Note the absolute scores: this benchmark is brutally hard and a 10% pass rate is the current state of the art, not a good result in plain English.
• DeepSWE v1.1 — Google now states in prose that 3.8 Flash clears "over 70% success" on this long-horizon software-engineering benchmark, ahead of most larger frontier models and "at a fraction of the cost." It still publishes a cost-versus-score chart rather than an exact figure, so treat 70% as the floor Google is willing to commit to, not a measured result.

Independent measurements tell a far more sober story — and this is where the first version of this page, along with most coverage published in September, was simply wrong. Artificial Analysis puts Gemini 3.8 Flash (high) at an Intelligence Index of 41 on Intelligence Index v4.3.2, ranked #50 of 224 models against a median of 26. Above average, solidly mid-field, not frontier. Output speed is 239.1 tokens per second, time to first token is 13.45 seconds, and cost per Intelligence Index task is $1.24.

Two things to know before you quote any of those numbers back at someone.

Always check the index version. Artificial Analysis rebased the Intelligence Index to v4.3.2 in October 2026. Under the previous v4.1.1 index this same model was widely quoted at 59 — the figure that circulated through every launch write-up in September, including ours. There is no conversion factor between the two versions and the scores are not comparable. An Intelligence Index number without a version stamp next to it is unusable, and that is exactly how a 59 and a 41 end up describing the same model.

The separate Agentic and Coding indices are gone. v4.3.2 is one weighted composite of ten evaluations: Agents 30% (AA-Briefcase v1.1, GDPval-AA v2.1, AutomationBench-AA), General 30% (AA-Omniscience, GDP.pdf, AA-LCR v1.1), Coding 20% (Terminal-Bench 4.0, SciCode) and Scientific Reasoning 20% (Humanity's Last Exam, CritPt). If you are still ranking models on a standalone "agentic index" figure — this page used to quote 50.0 for 3.8 Flash against 45.1 for 3.7 Flash — that composite was retired, and no v4.3.2 replacement exists for it. We have dropped those rows rather than restate v4.1.1 numbers alongside v4.3.2 ones.

On LMArena's text leaderboard, Gemini 3.8 Flash High sits at #8 with 1495 ±5 over 26,298 votes, still marked preliminary — a single point clear of Claude Opus 4.7 High at 1494 ±4. Gemini 3.7 Flash High is #17 at 1488 ±5. Blind human preference is the one axis where this tier genuinely trades punches with last generation's flagships, and it is worth more weight than it usually gets precisely because it is not a benchmark anyone can rebase.

The fair summary is less flattering than Google's slides suggest. Gemini 3.8 Flash is a mid-field model on independent reasoning benchmarks that wins decisively on one axis: throughput. Its closest open-weight rival makes the point uncomfortably well. GLM-5.3-Flash scores 42 on the same Intelligence Index v4.3.2 — one point above Gemini 3.8 Flash — at $0.15/$0.50 per 1M tokens and $0.25 per Intelligence Index task against Gemini's $1.24. Roughly five times the cost, for marginally less measured intelligence.

So the honest division of labour: GLM-5.3-Flash is the value pick, Gemini 3.8 Flash is the throughput pick. Gemini streams about 4.6× faster once it gets going, which matters enormously behind a user-facing stream and not at all for an overnight queue. If your workload is a queue rather than a chat window, the cheaper model currently wins on both measured intelligence and bill.

What Gemini keeps uncontested in this bracket is the input surface — a megatoken of context plus native video and audio ingestion, covered in the next section. That, not the index score, is the defensible reason to pay the premium.

What can Gemini 3.8 Flash actually take as input?

This is where the Flash tier genuinely separates from its price-matched rivals. Most models at $0.75 per million input tokens take text and images. Gemini 3.8 Flash takes video and audio too, at meaningful lengths.

Run the arithmetic and something clicks: 3 hours of low-res video at 100 tokens/second is 1,080,000 tokens. One hour of high-res video at 300 tokens/second is also 1,080,000. And 9.5 hours of audio at 32 tokens/second is 1,094,400. All three caps are the 1M context window expressed in different units. They are not arbitrary product limits — they are the context ceiling, and they move together with it.

What that costs in practice, at the standard input rate:

• One hour of audio — 115,200 tokens, about $0.086 of input. Transcribing, summarising or extracting action items from an hour-long recording costs under nine cents before output.
• One hour of low-resolution video — 360,000 tokens, about $0.27 of input.

Per Google's audio documentation, incoming audio is downsampled to 16 Kbps and multi-channel audio is combined to a single channel, so don't expect speaker separation from the raw signal. Thirteen MIME types are accepted, including WAV, MP3, FLAC, AAC, Opus and M4A, and the model understands non-speech sound — birdsong, sirens — not just transcription.

On the video side, nine container formats are supported and public YouTube URLs can be passed directly. The free tier caps YouTube ingestion at 8 hours per day; the paid tier has no length-based limit. Gemini 3.8 Flash also supports agentic video understanding, added to the Flash line on 1 September 2026 — instead of statically tokenising every frame, the model navigates the timeline and requests transcripts, frames or audio tracks on demand, which Google says uses up to 88% fewer tokens on long-form content. On a three-hour video that is the difference between a $0.81 input bill and roughly a tenth of it.

How do you use Gemini 3.8 Flash?

The model string is gemini-3.8-flash. There is one stable version and no preview or experimental alias.

A minimal call against the Gemini API:

curl -s "https://generativelanguage.googleapis.com/v1beta/models/gemini-3.8-flash:generateContent" \
  -H "x-goog-api-key: $GEMINI_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "contents": [
      { "parts": [ { "text": "Summarise this changelog into five bullets." } ] }
    ]
  }'

In Python, with the google-genai SDK, dialling thinking down for a cheap extraction task:

from google import genai
from google.genai import types

client = genai.Client()  # reads GEMINI_API_KEY

resp = client.models.generate_content(
    model="gemini-3.8-flash",
    contents="Extract every version number from this changelog as JSON.",
    config=types.GenerateContentConfig(
        thinking_config=types.ThinkingConfig(thinking_level="low"),
        max_output_tokens=1024,
    ),
)

print(resp.text)
print(resp.usage_metadata)  # inspect thought tokens before you scale up

Check usage_metadata on a representative sample of real traffic before committing to a budget. The thinking-token count decides your bill, and it varies far more by task than input length does.

Where you can run it:

• Google AI Studio — free tier available; the rate-limit documentation confirms 3.8 Flash is live for free-tier projects. Per-tier RPM/TPM figures are shown inside AI Studio rather than in the public docs.
• Vertex AI / Gemini Enterprise Agent Platform — for production workloads that need enterprise controls and regional deployment.
• OpenRouter — as google/gemini-3.8-flash, with google/gemini-3.8-flash:batch at exactly half price. OpenRouter exposes six endpoints (AI Studio and Vertex global, each in standard, flex and priority variants), and maps Google's thinking_level onto the OpenAI-style reasoning_effort parameter. Supported parameters there include reasoning_effort, structured_outputs, seed, tools, tool_choice, response_format, temperature and top_p.

Should you upgrade from Gemini 3.7 Flash?

Not automatically — and that is a reversal of how this upgrade was first reported, including on this page. The list price is identical, so the migration looks free, but the independent measurements say you pay for the quality gain in latency and in thinking tokens. The technical migration really is just a model-string change; the question is whether you want the trade.

The decision rule, honestly stated:

• Upgrade if your workload is agentic, multi-step or long-horizon and runs asynchronously. That is where the gain sits — all three benchmarks Google published are agent benchmarks — and a 2.8-second first-token penalty costs a background job nothing.
• Stay on 3.7 Flash if you are latency-sensitive. 3.8 Flash is 2.8 seconds slower to first token and 15% slower to stream. There is no throughput gain offsetting the latency penalty this time; both moved the wrong way. Behind a streaming chat UI that is a plain regression.
• Stay on 3.7 Flash if you are cost-sensitive at acceptable quality. Two index points cost 33% more per completed task, because 3.8 Flash deliberates harder by default. Measure total_thought_tokens on both against your own traffic before deciding the two points are worth a third more spend.
• Price up GLM-5.3-Flash instead if the job is batchable. It scores a point higher on the same v4.3.2 index at about a fifth of the cost per task. The only thing you give up is speed, and in a queue speed is nearly free.
• Stay on 3.7 Flash if you have front-end design output you have already tuned and validated. It scores marginally higher on Design Arena's website category, and there is no cost saving to be had from moving.

Gemini 3.7 Flash is not deprecated. Google's model list still shows it as stable, labelled the "previous-generation Flash model," with no shutdown date announced. There is no forced-migration clock, which is exactly why you can spend a week benchmarking on your own traffic rather than trusting anyone's index.

The most important number in this launch is not a benchmark at all. It is 1 January 2027, when promotional pricing ends and every tier doubles: standard to $1.50/$7.50, batch and flex to $0.75/$3.75, priority to $2.70/$13.50, cached reads to $0.15 per 1M and cache storage to $1.00 per 1M per hour. The discounts double alongside the list price, so the relative savings survive — the absolute bill does not. Whatever you build on this tier, build the batch path and the caching path in now. They are the two levers that still work after the price change, and they are what makes a 33% higher cost per task survivable.

For where the Flash tier sits in Google's wider lineup, see our Gemini 3.5 complete guide. If you are cost-shopping this bracket, the closest open-weight rival is covered in our GLM-5.3-Flash guide.

FAQ

What is Gemini 3.8 Flash?

Gemini 3.8 Flash is Google's newest Flash-tier AI model, generally available on 2 September 2026 under the model ID gemini-3.8-flash. It accepts text, image, video, audio and PDF input, returns text only, has a 1,048,576-token context window with a 65,536-token output limit, and reasons by default at a configurable low/medium/high thinking level.

How much does Gemini 3.8 Flash cost?

$0.75 per 1M input tokens and $3.75 per 1M output tokens on the standard tier. The Batch API and Flex inference halve that to $0.375 and $1.875; Priority inference raises it to $1.35 and $6.75. Cached input reads cost $0.075 per 1M. These are promotional rates that double on 1 January 2027.

Is Gemini 3.8 Flash better than 3.7 Flash?

Smarter, but slower and dearer per task. On Artificial Analysis's Intelligence Index v4.3.2 it scores 41 against 3.7 Flash's 39, and it wins all three benchmarks Google published. Against that: output speed drops from 279.7 to 239.1 tokens per second, time to first token rises from 10.62 to 13.45 seconds, and cost per Intelligence Index task rises from $0.93 to $1.24 despite an identical per-token price.

What is Gemini 3.8 Flash's Artificial Analysis score?

41 on Artificial Analysis's Intelligence Index v4.3.2, ranked #50 of 224 models against a median of 26, as of October 2026. Always check the version. Under the older v4.1.1 index the same model was quoted at 59, the two numbers are not comparable, and Artificial Analysis published no conversion factor between them.

Is GLM-5.3-Flash better than Gemini 3.8 Flash?

On measured intelligence and cost, yes. GLM-5.3-Flash scores 42 on the same Intelligence Index v4.3.2 against Gemini 3.8 Flash's 41, at $0.15/$0.50 per 1M tokens and $0.25 per Intelligence Index task versus Gemini's $1.24. Gemini wins decisively on speed — 239.1 against 51.4 tokens per second — plus a megatoken context with native video and audio input.

What is Gemini 3.8 Flash's context window?

1,048,576 input tokens — one full megatoken — with a separate 65,536-token maximum output. This is identical to Gemini 3.7 Flash. The context ceiling is also what caps every other modality: 3 hours of low-resolution video, 1 hour of high-resolution video and 9.5 hours of audio all work out to roughly 1M tokens.

Does Gemini 3.8 Flash support audio and video?

Yes, as input. It accepts up to 9.5 hours of audio per prompt at 32 tokens per second, and video up to 3 hours at low media resolution or 1 hour at high, plus public YouTube URLs. It also supports agentic video understanding, which Google says cuts token use by up to 88% on long-form content. It cannot generate audio or video.

Is Gemini 3.8 Flash free?

There is a free tier in Google AI Studio, and Gemini 3.8 Flash is available on it — with rate limits, and with an 8-hour-per-day cap on YouTube video ingestion. Exact free-tier RPM and TPM figures are shown inside AI Studio rather than published in the docs. Production use runs on the paid tiers above.

Is Gemini 3.7 Flash deprecated?

No. Google's model list still shows gemini-3.7-flash as a stable model, now labelled the previous-generation Flash model, with no announced shutdown date. The only near-term Gemini deprecation on the changelog is gemini-omni-flash-preview, retiring 30 September 2026. You can migrate to 3.8 Flash on your own schedule.

Do thinking tokens cost extra on Gemini 3.8 Flash?

They are billed at the output rate of $3.75 per 1M, and Google's docs state response pricing is the sum of output and thinking tokens. Because thinking is on by default at medium, real costs routinely run around twice a naive estimate based on visible output alone. Check total_thought_tokens in usage metadata, and drop thinking_level to low where deliberation isn't needed.