Models · Google (Gemini / DeepMind) · Out sinceReleased 21 Jul 2026

Gemini 3.5 Flash-Lite

Available on OpenRouterOn OpenRouterAvailable nowGenerally availableClosed (can't be downloaded)Proprietary

Where you can use it:In apps:Gemini in Google Workspace · Flash-Lite

Gemini 3.5 Flash-Lite is made by Google (Gemini / DeepMind). We don't have enough test results yet to rank it. It's cheap to use.

Current cheap tier. API default thinking_level = minimal; Google's published evals use high thinking.

Writing & creativity: —
Research & analysis: —
Coding: —
Price: Cheap$0.30 / $2.50
Price per 1M (blended)Blended / 1M: $0.85
MemoryContext: 1M
Longest answerMax output: 66K
Test resultsResults: 10 (1 independent1 indep.)
Out sinceReleased: 21 Jul 2026
Made byVendor: Google (Gemini / DeepMind)
UnderstandsInputs: text, image, audio, video

Scores are out of 100 for each area, compared with every model we track; “#” is its rank. Context is how much text it can read at once (1M is roughly 700,000 words). Blended price mixes the cost of what you send and what it writes back.

Thinking levels

Reasoning settings

MinimalLowMediumHigh

Links

Using it through OpenRouter

On OpenRouter

OpenRouter is a service that gives access to many AI models in one place. This model is listed there as google/gemini-3.5-flash-lite.

Route it as google/gemini-3.5-flash-lite at $0.30 in / $2.50 out per 1M tokens, 1M context. Listed since 21 Jul 2026.

include_reasoningmax_tokensreasoningreasoning_effortresponse_formatseedstopstructured_outputstemperaturetool_choicetoolstop_p

Every test result

Every result

All the numbers,with where they came from.

Every result we've found for this model, with who measured it and a link to where we read it.

All raw rows (vendor and independent kept separate), with setting label, source, date, harness and notes.

All results for this model
						NotesNotes
CharXiv Reasoning (no tools)	74.5%	Highhigh thinking	Maker's own figureVendor-reportedGoogle ↗	21 Jul 2026	—
CharXiv Reasoning (with tools)	76.5%	Highhigh thinking	Maker's own figureVendor-reportedGoogle ↗	21 Jul 2026	—
EuroEval Swedish (generative)	1.7	Defaultgemini/gemini-3.5-flash-lite (zero-shot, val)	Independent testIndependentEuroEval (Alexandra Institute) ↗	29 Sep 2026	—	EuroEval rank tier 5; ±0.06; lower is better; task scores (first metric): SweDN summarisation 36.72 ± 0.20, Skolprov 64.22 ± 2.47, Swedish facts 62.26 ± 2.61, ScaLA-sv 70.68 ± 1.13
GDPval-AA v2	1140	Highhigh thinking	Maker's own figureVendor-reportedGoogle ↗	21 Jul 2026	—
MLE-Bench (Partial 30)	39.2%	Highhigh thinking	Maker's own figureVendor-reportedGoogle ↗	21 Jul 2026	interactive Bash harness
MRCR v2 (8-needle, 1M pointwise)	21.3%	Highhigh thinking	Maker's own figureVendor-reportedGoogle ↗	21 Jul 2026	—	1M pointwise.
OpenAI MRCR v2 (8-needle)	72.2%	Highhigh thinking	Maker's own figureVendor-reportedGoogle ↗	21 Jul 2026	—	128k average.
OSWorld-Verified	74.0%	Highhigh thinking	Maker's own figureVendor-reportedGoogle ↗	21 Jul 2026	—	Avg of 5 runs.
SWE-Bench Pro (public, v1)	54.2%	Highhigh thinking	Maker's own figureVendor-reportedGoogle ↗	21 Jul 2026	internal Antigravity harness	Full public set.
Terminal-Bench 2.1	54.0%	Highhigh thinking	Maker's own figureVendor-reportedGoogle ↗	21 Jul 2026	Terminus 2