Google: Gemini 2.5 Flash Lite

by Google

Gemini 2.5 Flash-Lite is a lightweight reasoning model in the Gemini 2.5 family, optimized for ultra-low latency and cost efficiency. It offers improved throughput, faster token generation, and better performance across common benchmarks compared to earlier Flash models. By default, "thinking" (i.e. multi-pass reasoning) is disabled to prioritize speed, but developers can enable it via the [Reasoning API parameter](https://openrouter.ai/docs/use-cases/reasoning-tokens) to selectively trade off cost for intelligence.

Avg Score

74.8%

285 answers

Avg Latency

1.8s

44 runs

Pricing

$0.10

input

$0.40

output

per 1M tokens

Context

1049K

tokens

Alternatives

Models with similar or better quality but different tradeoffs

Same Quality, Cheaper

Models with similar or better performance at a lower cost per token.

Model	Cost
Mistral: Devstral 2 2512	-100%
Xiaomi: MiMo-V2-Flash (free)	-100%
Meta: Llama 3.3 70B Instruct	-100%
Google: Gemma 3 4B	-100%
Google: Gemma 3 27B	-100%

Same Quality, Faster

Models with similar or better performance but lower latency.

Model	Latency
Google: Gemini 2.0 Flash	-27%
OpenAI: gpt-oss-120b	-20%
Google: Gemini 2.5 Flash	-16%
Anthropic: Claude 3 Haiku	-5%

Same Cost, Better

Models at a similar price point with higher benchmark scores.

Model	Score
Mistral: Pixtral 12B	+15%
Nous: Hermes 4 70B	+15%
Google: Gemma 3 12B	+15%
Xiaomi: MiMo-V2-Flash (free)	+15%
OpenAI: GPT-4.1 Nano	+15%

Other Models from Google

Compare performance with other models from the same creator

Model	Score	Latency	Cost/1M
Google: Gemini 3 Pro Preview	91.5%	10.8s	$7.00
Google: Gemini 3 Flash Preview	90.1%	2.6s	$1.75
Google: Gemini 2.5 Pro Preview 06-05	90.0%	17.0s	$5.63
Google: Gemma 3 12B	90.0%	8.0s	$0.0000
Google: Gemma 3 12B	90.0%	9.5s	$0.07
Google: Gemini 2.5 Pro	87.5%	10.1s	$5.63
Google: Gemma 3 27B	85.0%	15.7s	$0.09
Google: Gemini 2.5 Flash	80.2%	2.5s	$1.40
Google: Gemini 2.5 Flash Lite Preview 09-2025	80.0%	2.2s	$0.25
Google: Gemma 3 4B	80.0%	9.8s	$0.0000
Google: Gemma 3 27B	80.0%	7.7s	$0.0000
Google: Gemma 2 27B	80.0%	5.6s	$0.65
Google: Gemini 2.0 Flash	79.2%	2.4s	$0.25
Google: Gemini 2.5 Flash Image (Nano Banana)	70.0%	2.0s	$1.40
Google: Gemini 2.0 Flash Lite	70.0%	2.1s	$0.19
Google: Nano Banana Pro (Gemini 3 Pro Image Preview)	60.0%	61.2s	$7.00
Google: Gemma 3n 4B	60.0%	8.5s	$0.0000
Google: Gemma 3 4B	60.0%	7.9s	$0.04
Google: Gemini 2.5 Flash Preview 09-2025	50.0%	3.0s	$1.40
Google: Gemma 3n 2B (free)	50.0%	10.2s	$0.0000
Google: Gemma 3n 4B	50.0%	6.6s	$0.03
Google: Gemini 2.5 Pro Preview 05-06	50.0%	16.5s	$5.63
Google: Gemma 2 9B	40.0%	2.0s	$0.06
Google: Gemini 2.5 Flash Image Preview (Nano Banana)	—	—	$1.40
Google: Gemini 2.0 Flash Experimental (free)	—	—	$0.0000

Benchmark Performance

How this model performs across different benchmarks

Benchmark	Score	Rank
moose	100.0%	11 / 17
Math Bench: Subtraction (1-10)	100.0%	17 / 35
Math Bench: Division (1-10)	100.0%	22 / 35
Single Character Instruction Adherence	100.0%	7 / 17
Breakfast	100.0%	10 / 17
my test	100.0%	7 / 17
Math Bench: Arithmetic (1-100)	100.0%	18 / 35
Customer service	100.0%	8 / 17
Venture Capital Terms Benchmark	100.0%	11 / 25
Best bathroom technique	100.0%	8 / 17
Repetitive Onomatopoeia Response Benchmark	100.0%	7 / 17
Math Bench: Addition (1-10)	100.0%	15 / 35
Customer support	100.0%	4 / 17
Instruction Adherence and Logical Constraints	100.0%	9 / 17
Cheeky_test	100.0%	8 / 17
Customer Support Ticket Generation	100.0%	8 / 17
Math Bench: Multiplication (1-10)	100.0%	17 / 35
Structured Customer Support Extraction	100.0%	9 / 17
Smell test	100.0%	7 / 17
SaaS Customer Support Extraction and Strategy	100.0%	9 / 17
Motor production test	95.0%	25 / 36
Maak een erp systeem	95.0%	3 / 17
Spatial Reasoning: Germany	94.3%	14 / 35
Product Recommendations	93.3%	6 / 17
Enhancing FIDO UV Consistency with UXWG 2026 priority plan	90.0%	16 / 25
KUKA force Torque control programming skills	90.0%	10 / 17
Westat Test	90.0%	8 / 292
Math Bench: Arithmetic (1-1000)	90.0%	23 / 35
Omnishambles	85.0%	9 / 17
Karlsruhe Local Knowledge Benchmark	66.2%	25 / 35
cemantle solve initiator	60.0%	22 / 25
Product Recommendations	52.5%	32 / 36
Rust performance	50.0%	6 / 16
Character Frequency Bench	41.8%	32 / 35
Testing_website	40.0%	16 / 17
solve n-queens in clojure in one shot	40.0%	10 / 17
Define data taxonomy for product catalog suitable for PIM	40.0%	17 / 17
Money Boy Cultural Literacy Test	28.6%	21 / 35
Docs	28.3%	11 / 17
German Memelord Bench	7.6%	28 / 35
Karl Lorey knowlege	0.0%	30 / 36
Current Global Political Leaders	0.0%	6 / 17

Price vs Performance

Compare cost efficiency across all models

Current model

Other models

X-axis uses log scale for better visualization of price range

Score Over Time

Performance trends across all benchmark runs

Benchmark Activity

Number of benchmark runs over time

Quickstart

Get started with this model using OpenRouter

View on OpenRouter

import { OpenRouter } from "@openrouter/sdk";

const openrouter = new OpenRouter({
  apiKey: "<OPENROUTER_API_KEY>"
});

const completion = await openrouter.chat.completions.create({
  model: "google/gemini-2.5-flash-lite",
  messages: [
    {
      role: "user",
      content: "Hello!"
    }
  ]
});

console.log(completion.choices[0].message.content);

Get your API key at openrouter.ai/keys