Models / Gemini 2.5 Flash

Google

Gemini 2.5 Flash

Gemini 2.5 Flash is strongest at sticking to the facts; worth a look for simple, high-volume work where cost matters most. The cheapest model of similar intelligence is gpt-oss-20b, at $0.06 per million tokens against $0.85.

Superseded. Gemini 3.8 Flash is Google's newer model in this line, at $3.00 blended: see Gemini 3.8 Flash or compare the two.

Retiring 1 Dec 2026. Google retires gemini-2.5-flash; the official replacement is gemini-3.5-flash-lite. How to re-test before you switch.

Results 16 Nov 2026 · 40 results · 5 sources

Price per million tokens
$0.30 in · $2.50 out
Google's list price, 10 Nov 2026 · compare prices$0.03 cached input
Speed
Not measured
Intelligence Index
13.1
Artificial Analysis, at thinking reasoning; 9.9–13.1 across 2 settings
Spring Prompt overall
Not ranked yet
needs our benchmarks and two groups of results
Developer
Google, based in the US
Weights not published
Released
17 Apr 2025
1,048,576 tokens of context

Where it stands among the models you could choose

Available models grouped by price, intelligence, speed and developer location. Retired models are excluded.

AmongIntelligence IndexSpring Prompt overallPrice, blendedSpeed
Models available today 119th of 185Claude Opus 5.5 – $0.85 blended, 52nd percentilecheapest: Mistral Nemo –
Models priced $0.25–1 per million tokens 34th of 61MiMo-V2.6-Pro – sets this group –
Models with similar intelligence sets this group – $0.85 blended, 74th percentilecheapest: gpt-oss-20b –
Models from developers based in the US 64th of 92Claude Opus 5.5 – $0.85 blended, 38th percentilecheapest: Gemma 3 4B –

The name under each rank is the leader of that group; green is the top quarter of the group and red the bottom quarter, by rank. Price is shown as a percentile: the share of the group that costs less, so lower is cheaper. "Sets this group" marks the measure the group is defined by. Spring Prompt overall is our score out of 100 across every source (how it works). Price is the developer's list price (or the typical OpenRouter host where we haven't read one) for three input tokens to one output; intelligence and speed are from Artificial Analysis (data sourced from Artificial Analysis); "similar intelligence" means within 4 points on its Intelligence Index. Developer locations are where each company is based, not where a model is served. Groups under 3 models are not ranked.

How good is it, and for what?

Each use case ranks the available models its benchmarks measured, then those priced $0.25–1 per million tokens. The bar shows its percentile in that field, best to the right. Open a row for the results behind it.

Product listingsTurning a sparse product feed and photos into listings that can go live Not measured
  • CatalogBench: has not measured this model.
Decks from an analysisTurning a finished analysis into a deck you could present as it is Not measured
  • DeckBench: has not measured this model.
User surveysPlanning a user survey and reading its results without being misled Not measured
  • SurveyBench: has not measured this model.
Marketing planningPlanning a year of ad spend without overspending Not measured
  • ROASBench: has not measured this model.
Agents and tool useMulti-step tasks with tools: support desks, coding agents, function calls 82nd of 159 available · 18th of 47 at its price
BenchmarkGemini 2.5 FlashRank among availableBest available
Berkeley Function Calling Leaderboard (BFCL) V4BFCL: overall accuracy · reported 56.2% 8th of 36 Claude Opus 4.5 77.5%
Vending-Bench 2Vending-Bench 2: money after a year · reported $548.84 44th of 53 GPT-6 Astra $15,515
  • tau2-bench: has not measured this model.
  • OpenHands Index: has not measured this model.
  • Microsoft STATE-Bench: has not measured this model.
  • Artificial Analysis: has not measured this model.
Professional workReal tasks from banking, consulting and law, business documents and freelance projects Not measured
  • APEX-Agents: has not measured this model.
  • GDP.pdf: has not measured this model.
  • Remote Labor Index: has not measured this model.
Reasoning and knowledgeHard questions across science, maths and general knowledge 121st of 187 available · 35th of 61 at its price
BenchmarkGemini 2.5 FlashRank among availableBest available
Artificial AnalysisArtificial Analysis: Artificial Analysis Intelligence Index · reported 13.1 119th of 185 Claude Opus 5.5 57.6
Artificial AnalysisArtificial Analysis: Humanity's Last Exam · reported 12.1% 121st of 186 Claude Opus 5.5 61.4%
Artificial AnalysisArtificial Analysis: GPQA Diamond · reported 79.0% 98th of 168 GPT-6 Astra 96.3%
WritingWhat people prefer in blind comparisons, and judged writing quality 96th of 169 available · 24th of 58 at its price
BenchmarkGemini 2.5 FlashRank among availableBest available
Arena (formerly LMArena)Arena Text: overall · reported 1,409 91st of 153 Gemini 4 Argon 1,525
  • UGI Leaderboard: has not measured this model.
Sticking to the factsSummarising without inventing things, and factual answers 65th of 139 available · 11th of 40 at its price
BenchmarkGemini 2.5 FlashRank among availableBest available
Vectara Hallucination LeaderboardVectara: hallucination rate · reported 7.8% 27th of 71 GPT-5.4 nano 3.1%
Arena (formerly LMArena)Arena Text factuality: overall · reported 1,437 68th of 109 Gemini 4 Argon 1,511
  • SimpleQA Verified (Epoch AI): has not measured this model.
Speed under pressureGood decisions against a real clock (fast chess) Not measured
  • BulletBench: has not measured this model.

Nearest alternatives

Against Gemini 2.5 Flash at $0.85 blended, Intelligence Index 13.1. "Similar intelligence" means within 4 points or better.

Where to run it

1 host on OpenRouter at the standard service tier, with a median of $0.85 blended per million tokens. Cheapest first.

HostQuantisationInputOutputBlended
Google AI Studionot stated$0.30$2.50$0.85

Other service tiers, not counted above: Google AI Studio flex $0.15 in, $1.25 out; Google AI Studio priority $0.54 in, $4.50 out. Flex and batch are cheaper but slower or queued; priority and fast cost more.

Prices per million tokens from OpenRouter's public endpoint list, checked 10 Nov 2026.

Against the alternatives

The models you would most likely weigh it against: the leaders of the groups above.

Scroll sideways to see every alternative.

Gemini 2.5 FlashMiMo-V2.6-Pro
most intelligent at $0.25–1 per million tokens
gpt-oss-20b
cheapest with similar intelligence
Claude Opus 5.5
most intelligent available today
Price per million tokens $0.85 $0.54$0.06$8.00
Tokens a second – 4327796
Intelligence Index 13.1 46.310.057.6
Spring Prompt overall – ––82
Rank among available models, by use case
Product listings – ––13th
Decks from an analysis – ––7th
User surveys – ––4th
Marketing planning – ––9th
Agents and tool use 82nd of 159 23rd130th3rd
Professional work – ––3rd
Reasoning and knowledge 121st of 187 13th132nd1st
Writing 96th of 169 23rd155th5th
Sticking to the facts 65th of 139 10th–3rd
Speed under pressure – –––

Shaded figures are better than Gemini 2.5 Flash's. A dash means no result.

Research and examples

Where it has been measured

34th most intelligent of the 61 models priced $0.25–1 per million tokens. Among models available today, mid-field on every use case measured so far.

Artificial Analysis lists it but has not published a speed for it yet

Every result

40 published results from 5 sources, each in the source's own units, with the configuration that produced it.

Arena (formerly LMArena) · reported by Arena (formerly LMArena) · 7 results
MeasureValueRankConfigurationDated
Arena Text factuality: overallArena rating, higher is better 1,437
1,434–1,439
99 of 182 Gemini 2.5 Flash 2 Oct 2026
Arena Text: business, management and financeArena rating, higher is better 1,399
1,395–1,404
163 of 406 Gemini 2.5 Flash 2 Oct 2026
Arena Text: creative writingArena rating, higher is better 1,394
1,389–1,399
117 of 411 Gemini 2.5 Flash 2 Oct 2026
Arena Text: expert promptsArena rating, higher is better 1,424
1,417–1,430
151 of 363 Gemini 2.5 Flash 2 Oct 2026
Arena Text: instruction followingArena rating, higher is better 1,399
1,396–1,403
145 of 413 Gemini 2.5 Flash 2 Oct 2026
Arena Text: overallArena rating, higher is better 1,409
1,407–1,412
148 of 413 Gemini 2.5 Flash 2 Oct 2026
Arena Text: writing, literature and languageArena rating, higher is better 1,402
1,398–1,406
123 of 412 Gemini 2.5 Flash 2 Oct 2026

Not shown: Not an error rate: claims that cannot be checked on the web are skipped, and preference still carries most of the weight.

Artificial Analysis · reported by Artificial Analysis · 14 results
MeasureValueRankConfigurationDated
Artificial Analysis: Artificial Analysis Intelligence Indexindex score, higher is better 9.9 320 of 413 Gemini 2.5 Flash (no reasoning) 20 Oct 2026
Artificial Analysis: Artificial Analysis Intelligence Indexindex score, higher is better 13.1 276 of 413 Gemini 2.5 Flash (thinking reasoning) 20 Oct 2026
Artificial Analysis: GPQA Diamond% of questions, higher is better 68.3% 256 of 360 Gemini 2.5 Flash (no reasoning) 20 Oct 2026
Artificial Analysis: GPQA Diamond% of questions, higher is better 79.0% 181 of 360 Gemini 2.5 Flash (thinking reasoning) 20 Oct 2026
Artificial Analysis: Humanity's Last Exam% of questions, higher is better 4.7% 352 of 419 Gemini 2.5 Flash (no reasoning) 20 Oct 2026
Artificial Analysis: Humanity's Last Exam% of questions, higher is better 12.1% 255 of 419 Gemini 2.5 Flash (thinking reasoning) 20 Oct 2026
Artificial Analysis: IFBench% of instructions, higher is better 39.0% 212 of 288 Gemini 2.5 Flash (no reasoning) 20 Oct 2026
Artificial Analysis: IFBench% of instructions, higher is better 50.3% 140 of 288 Gemini 2.5 Flash (thinking reasoning) 20 Oct 2026
Artificial Analysis: Long-context reasoning (AA-LCR)% of questions, higher is better 49.9% 282 of 407 Gemini 2.5 Flash (no reasoning) 20 Oct 2026
Artificial Analysis: Long-context reasoning (AA-LCR)% of questions, higher is better 65.3% 229 of 407 Gemini 2.5 Flash (thinking reasoning) 20 Oct 2026
Artificial Analysis: Terminal-Bench Hard% of tasks, higher is better 12.1% 190 of 282 Gemini 2.5 Flash (no reasoning) 20 Oct 2026
Artificial Analysis: Terminal-Bench Hard% of tasks, higher is better 13.6% 182 of 282 Gemini 2.5 Flash (thinking reasoning) 20 Oct 2026
Artificial Analysis: τ²-bench telecom% of tasks, higher is better 14.9% 272 of 286 Gemini 2.5 Flash (no reasoning) 20 Oct 2026
Artificial Analysis: τ²-bench telecom% of tasks, higher is better 31.6% 206 of 286 Gemini 2.5 Flash (thinking reasoning) 20 Oct 2026

Not shown: Not business work, and a blend: read the parts for any one task.

Berkeley Function Calling Leaderboard (BFCL) V4 · reported by Berkeley Function Calling Leaderboard (BFCL) V4 · 16 results
MeasureValueRankConfigurationDated
BFCL: irrelevance detection% correct, higher is better 93.7% 6 of 109 Gemini 2.5 Flash (bfcl fc) 16 Dec 2025
BFCL: irrelevance detection% correct, higher is better 91.1% 11 of 109 Gemini 2.5 Flash (bfcl prompt) 16 Dec 2025
BFCL: memory% correct, higher is better 41.3% 17 of 109 Gemini 2.5 Flash (bfcl fc) 16 Dec 2025
BFCL: memory% correct, higher is better 38.7% 18 of 109 Gemini 2.5 Flash (bfcl prompt) 16 Dec 2025
BFCL: multi-turn tasks% correct, higher is better 36.3% 32 of 109 Gemini 2.5 Flash (bfcl fc) 16 Dec 2025
BFCL: multi-turn tasks% correct, higher is better 16.8% 53 of 109 Gemini 2.5 Flash (bfcl prompt) 16 Dec 2025
BFCL: overall accuracy% correct, higher is better 56.2% 15 of 109 Gemini 2.5 Flash (bfcl fc) 16 Dec 2025
BFCL: overall accuracy% correct, higher is better 50.9% 26 of 109 Gemini 2.5 Flash (bfcl prompt) 16 Dec 2025
BFCL: relevance detection% correct, higher is better 75.0% 62 of 109 Gemini 2.5 Flash (bfcl fc) 16 Dec 2025
BFCL: relevance detection% correct, higher is better 62.5% 91 of 109 Gemini 2.5 Flash (bfcl prompt) 16 Dec 2025
BFCL: single-turn calls (curated)% correct, higher is better 85.0% 44 of 109 Gemini 2.5 Flash (bfcl fc) 16 Dec 2025
BFCL: single-turn calls (curated)% correct, higher is better 88.1% 21 of 109 Gemini 2.5 Flash (bfcl prompt) 16 Dec 2025
BFCL: single-turn calls (user-contributed)% correct, higher is better 74.4% 45 of 109 Gemini 2.5 Flash (bfcl fc) 16 Dec 2025
BFCL: single-turn calls (user-contributed)% correct, higher is better 78.2% 27 of 109 Gemini 2.5 Flash (bfcl prompt) 16 Dec 2025
BFCL: web search% correct, higher is better 59.0% 21 of 109 Gemini 2.5 Flash (bfcl fc) 16 Dec 2025
BFCL: web search% correct, higher is better 62.0% 20 of 109 Gemini 2.5 Flash (bfcl prompt) 16 Dec 2025

Not shown: Not general refusal or safety behaviour.

Vectara Hallucination Leaderboard · reported by Vectara Hallucination Leaderboard · 2 results
MeasureValueRankConfigurationDated
Vectara: answer rate% of documents, higher is better 99.0% 68 of 108 Gemini 2.5 Flash 22 Sep 2026
Vectara: hallucination rate% of summaries, lower is better 7.8% 39 of 108 Gemini 2.5 Flash 22 Sep 2026

Not shown: Not a quality score: a low rate usually means content filters were triggered, and hallucination rates are measured on answered documents only.

Vending-Bench 2 · reported by Vending-Bench 2 · 1 result
MeasureValueRankConfigurationDated
Vending-Bench 2: money after a yearUS dollars, higher is better $548.84 52 of 63 Gemini 2.5 Flash 1 Oct 2026

Not shown: Not a real business; one simulated market with set rules.

Test your own prompt.

Benchmarks aren't your task. Run your prompt on the models you could use, with every answer checked and costed.