Migrating from gemini-3-flash-preview to gemini-3.5-flash

What changes when you move from gemini-3-flash-preview to gemini-3.5-flash: input price +200%, output price +200%.

From: gemini-3-flash-preview
Status: Active
Shutdown: none announced
Time left: —

To: gemini-3.5-flash
Status: Active
Shutdown: none announced
Time left: —

Prices

USD per 1M tokens, standard tier

Type	gemini-3-flash-preview	gemini-3.5-flash	Change
Input	$0.50first seen October 1, 2026 · source	$1.50first seen October 1, 2026 · source	+200%
Cached input	$0.05first seen October 1, 2026 · source	$0.15first seen October 1, 2026 · source	+200%
Output	$3.00first seen October 1, 2026 · source	$9.00first seen October 1, 2026 · source	+200%

At 1M input and 250K output tokens a month: $1.25 on gemini-3-flash-preview, $3.75 on gemini-3.5-flash (+200%).

Other tiers, prompt-size bands and modalities are on each model's page.

Limits

Tokens

Limit	gemini-3-flash-preview	gemini-3.5-flash	Change
Context window	1,048,576 tokensin place when tracking began (October 6, 2026) · source	1,048,576 tokensin place when tracking began (October 6, 2026) · source	0%
Max output tokens	65,536 tokensin place when tracking began (October 6, 2026) · source	65,536 tokensin place when tracking began (October 6, 2026) · source	0%

Benchmarks

Same benchmark and source on both sides; settings named where they differ

Benchmark	gemini-3-flash-preview	gemini-3.5-flash	Change
Arena TextArena, high effort (vs no setting). Other settings not rated → 1476 ±4 medium effort	1473 ±4	1477 ±4	+4 within range
Arena Text: CodingArena, high effort (vs no setting). Other settings not rated → 1506 ±6 medium effort	1506 ±8	1508 ±6	+2 within range
Arena WebDevArena, high effort (vs no setting). Ranges don't overlap. Other settings not rated → 1491 ±7 medium effort	1440 ±9	1499 ±7	+59
Arena VisionArena, medium effort (vs no setting). Other settings not rated → 1284 ±7 high effort	1273 ±5	1285 ±7	+12 within range
Arena DocumentArena, medium effort (vs no setting). Ranges don't overlap. Other settings not rated → 1453 ±9 high effort	1439 ±9	1461 ±8	+22
FrontierMath Tier 4Epoch AI, high effort (vs no setting).	17.1%	26.8%	+9.7 pts
FrontierMath Tiers 1-3Epoch AI, high effort (vs no setting).	51.2%	62.8%	+11.6 pts
GPQA DiamondEpoch AI, high effort. Other settings not run → 88.9% low effort not run → 86.4% minimal effort	89.4%	92.8%	+3.4 pts
OTIS Mock AIME 2024-2025Epoch AI, high effort. Other settings not run → 88.9% low effort not run → 80.0% minimal effort	95.6%	95.6%	0 pts
SimpleQA VerifiedEpoch AI, high effort.	66.8%	66.2%	−0.6 pts
SWE-bench VerifiedEpoch AI, high effort (vs no setting).	75.4%	79.3%	+3.9 pts

A score from one source is never compared with another source's, and company-reported scores are compared only within the same company; settings named where they differ. Data: Arena (arena.ai), CC BY 4.0; Epoch AI, Capabilities & benchmarking (epoch.ai/benchmarks), CC BY.

Features and parameters

Compare models

Same on both

Structured outputs: supported on both.gemini-3-flash-preview: in place when tracking began (October 6, 2026) · source · gemini-3.5-flash: in place when tracking began (October 6, 2026) · source
Function calling: supported on both.gemini-3-flash-preview: in place when tracking began (October 6, 2026) · source · gemini-3.5-flash: in place when tracking began (October 6, 2026) · source
Prompt caching: supported on both.gemini-3-flash-preview: in place when tracking began (October 6, 2026) · source · gemini-3.5-flash: in place when tracking began (October 6, 2026) · source
Batch API: supported on both.gemini-3-flash-preview: in place when tracking began (October 6, 2026) · source · gemini-3.5-flash: in place when tracking began (October 6, 2026) · source
Image input: supported on both.gemini-3-flash-preview: in place when tracking began (October 6, 2026) · source · gemini-3.5-flash: in place when tracking began (October 6, 2026) · source
PDF input: supported on both.gemini-3-flash-preview: in place when tracking began (October 6, 2026) · source · gemini-3.5-flash: in place when tracking began (October 6, 2026) · source
Reasoning / thinking: supported on both.gemini-3-flash-preview: in place when tracking began (October 6, 2026) · source · gemini-3.5-flash: in place when tracking began (October 6, 2026) · source
Built-in web search: supported on both.gemini-3-flash-preview: in place when tracking began (October 6, 2026) · source · gemini-3.5-flash: in place when tracking began (October 6, 2026) · source

Not published for gemini-3-flash-preview: Fine-tuning, temperature parameter, top_p parameter, logprobs.

Not published for gemini-3.5-flash: Fine-tuning, temperature parameter, top_p parameter, logprobs.

Free tier

gemini-3-flash-preview: Free tier, in place when tracking began (October 7, 2026) (source)

gemini-3.5-flash: Free tier, in place when tracking began (October 7, 2026) (source) · All free models

Aliases

All aliases

No aliases point at gemini-3-flash-preview.

Aliases pointing at gemini-3.5-flash: gemini-flash-latest (since May 19, 2026, per Google · source)

On cloud platforms

All platform dates

Model	Platform	Platform ID	Shutdown	Time left
gemini-3.5-flash	Google Cloud Vertex AI	gemini-3.5-flash	not before May 19, 2027Active

Read daily from each platform's own pages. The status above uses each company's own date. Time left counts to 00:00 UTC on the shutdown date, which is the evening before in ET.

Checklist

Change the model string from gemini-3-flash-preview to gemini-3.5-flash.
Re-check costs: gemini-3.5-flash costs more than gemini-3-flash-preview (input +200%, output +200%).

Compared from AIWire's published data when this page loaded. Each fact links to the page it came from. Model pages: gemini-3-flash-preview, gemini-3.5-flash.