Gemini 3.5 Flash benchmarks

Gemini 3.5 Flash is a proprietary model from Google, released 19 May 2026. It ranks 36th of 54 on our overall leaderboard and 25th of 41 for agents. At A$4.86 per million tokens (blended), it costs about the same as the median model we track. It can run with requests kept in Australia on Google Vertex AI in Sydney.

#36 OF 54 OVERALLA$4.86 / 1M TOKENS1M CONTEXTUPDATED 2 OCT 2026
CONTEXT WINDOW
1,048,576 tokens
MAX OUTPUT
65,536 tokens
INPUT
text, image, video, audio
WEIGHTS
Proprietary
RELEASED
19 May 2026
GOOGLE DOCS ↗
§ 02 — RESULTS

Benchmark results.

Independent test scores are for the high effort setting. “Among tracked” ranks Gemini 3.5 Flash against the 54 models on this site; LMArena ranks run across its full leaderboard.

ALL RESULTS7 OF 8 CORE SIGNALS
Gemini 3.5 Flash benchmark results
BENCHMARKRESULTAMONG TRACKEDSOURCE DETAIL
INDEPENDENT TESTS · ARTIFICIAL ANALYSIS
AA Intelligence Index32.6#40 of 54high effort
AA Coding Index70.1#25 of 42high effort
GPQA Diamond92.2%#21 of 44high effort
Humanity’s Last Exam42.7%#27 of 54high effort
AA-LCR73.3%#52 of 54high effort
Terminal-Bench 2.178.7%#24 of 42high effort
τ²-Bench (banking)32.2%#28 of 41high effort
SciCode53.9%#29 of 49high effort
BLIND HUMAN VOTES · LMARENA
LMArena Text47,981 votes · ±41477#24 of 49#27 of 410 on LMArena · 30 Sep 2026
LMArena WebDev9,495 votes · ±71499#43 of 52#53 of 137 on LMArena · 30 Sep 2026
LMArena Vision11,983 votes · ±71285#17 of 36#19 of 156 on LMArena · 28 Sep 2026
LMArena Document6,500 votes · ±81461#20 of 26#25 of 44 on LMArena · 13 Sep 2026
§ 03 — COST

What Gemini 3.5 Flash costs in Australian dollars.

List prices converted at the snapshot’s RBA rate, excluding GST. Reasoning models are billed for their thinking as output tokens, so heavier settings cost more than the job estimates show.

LIST PRICE
Gemini 3.5 Flash price per million tokens
PER MILLION TOKENSAUDUSD LIST
Input tokensA$2.16US$1.50
Output tokensA$12.95US$9.00
Blended (3 in : 1 out)A$4.86US$3.38
EVERYDAY JOBSESTIMATES
Estimated cost of common jobs in Australian dollars
JOBCOST (AUD)MEDIAN MODEL
Customer support replyPER 1,000 REPLIESA$8.20≈ A$8.20
Summarise a 30-page documentPER 100 DOCUMENTSA$5.35≈ A$5.35
Agentic coding taskPER 10 TASKSA$4.27≈ A$4.27
SETTINGSMEASURED SEPARATELY
Gemini 3.5 Flash settings compared
SETTINGINTELLIGENCEA$ / 1MSPEED
high effort32.6A$4.86
medium effort33.6A$4.86
minimal effort23.8A$4.86
§ 04 — AUSTRALIA

Running Gemini 3.5 Flash in Australia.

It can run with requests kept in Australia on Google Vertex AI in Sydney. How the platforms compare →

AUSTRALIAN CLOUD REGIONS
Gemini 3.5 Flash availability in Australian cloud regions
PLATFORMAVAILABILITYDETAIL
AWS Bedrock · SydneyNOT OFFERED
AWS Bedrock · MelbourneNOT OFFERED
Azure · Australia EastNOT OFFERED
Google Vertex AI · SydneyIN REGION

Appears to need Provisioned Throughput; pay-as-you-go access from Sydney goes through the global endpoint. Provider docs, checked 23 Sep 2026

§ 06 — QUESTIONS

Gemini 3.5 Flash, answered.

Gemini 3.5 Flash lists at US$1.50 per million input tokens and US$9.00 per million output tokens — A$2.16 and A$12.95 at A$1 = US$0.6949 (RBA, 1 Oct 2026), excluding GST. A typical customer support reply works out at about A$8.20 per 1,000 replies, before any reasoning tokens.

Gemini 3.5 Flash can run with requests kept in Australia on Google Vertex AI in Sydney. Availability comes from the cloud providers’ own documentation; check your provider’s current terms before relying on it for data-residency obligations.

It ranks 29th of 42 on our coding ranking, with an Artificial Analysis Coding Index of 70.1 and an LMArena WebDev rating of 1499 (53rd of 137). The current leader is Claude Fable 5.1.

It ranks 25th of 41 on our agents ranking, completing 32.2% of τ²-Bench banking customer-service tasks and 78.7% of Terminal-Bench 2.1 tasks.

Google documents a context window of 1,048,576 tokens (1M), with up to 65,536 output tokens. On AA-LCR, which tests reasoning across ~100,000-token document sets, it scores 73.3%.

PUT THE COMPARISON TO WORK

Need help choosing and using AI for your business?