modelbenchmark.io

Gemini 2.5 Flash

Google · gemini-2-5-flash

Compare

Fast Gemini workhorse for multimodal apps where latency and price matter

Specification

google listing

Context
1M
Max output
66K
Released
2025-06-17
Knowledge cutoff
2025-01
Retires
2026-10-02
Open weights
no
Input
text, image, audio, video, pdf
Output
text

Price

US dollars per million tokens · google list price

Input
$0.30
Output
$2.50
Cache read
$0.03
Cache write
$1.00

Batch and priority prices

US dollars per million tokens · per listing

PriceListing$/M
Batch inputvertex_ai-language-models$0.15
Batch outputvertex_ai-language-models$1.25
Priority inputvertex_ai-language-models$0.54
Priority outputvertex_ai-language-models$4.50

Batch input is 50% below the interactive input price at the same listing.

Rate limits

as listed

Requests per minute
100,000
Tokens per minute
8,000,000

Quality

2 benchmarks · 2 sources

Composite
37th
percentile of 344 scored models
Rank
216 / 344
±0.08 sd
Evidence
2 × 2
one source, or fewer than 3 benchmarks
Effort range
—
one configuration only
coding12th1/3 bench
math77th1/7 bench

Every score

one row per source and configuration — nothing averaged away

BenchmarkScoreConfigurationSourceRun
OTIS Mock AIME 2024-202571.9 ±5.7default · 2 runsEpoch AI2025-05-20
SWE-Bench verified28.7default · mini-SWE-agentSWE-bench—

The leading figure for a benchmark is the median across its configurations, so one heroic high-effort run cannot set the number.

Available from 38 hosts

HostIn $/MOut $/MCache rdCache wrContextOutputRetires
googlevendor$0.30$2.50$0.03—1M66K—
qihang-ai$0.09$0.71——1M66K—
kilo$0.15$1.25$0.015$0.04171M66K—
oci · oci$0.15$0.60——1M66K—
poe$0.21$1.80$0.021—1.1M66K—
jiekou$0.285$2.375$0.0285—1M66K—
cortecs$0.299$2.491$0.029$0.0971M66K—
302ai$0.30$2.50——1M66K—
abacus$0.30$2.50$0.03—1M66K—
aihubmix$0.30$2.499$0.03—1M66K—
auriko$0.30$2.50$0.03—1M66K—
cloudflare-ai-gateway$0.30$2.50$0.03—1M66K—
crossmodel$0.30$2.50$0.03$0.301M66K—
databricks$0.30$2.50$0.03$0.301M
66K
66K
2026-10-02
deepinfra$0.30$2.50——1M1M—
fastrouter$0.30$2.50$0.03—1M66K—
frogbot$0.30$2.50$0.075—1M66K—
gemini$0.30$2.50$0.03—1M66K—
google-vertex$0.30$2.50$0.03—1M66K—
helicone$0.30$2.50$0.075$0.301M66K—
impossibl$0.30$2.50$0.03—1M66K—
merge-gateway$0.30$2.50$0.03—1M66K—
modelis$0.30$2.50——1M66K—
nano-gpt$0.30$2.50$0.03—1M66K—
nearai$0.30$2.50$0.03—1M66K—
ofox$0.30$2.50$0.03$1.001M66K—
openrouter$0.30$2.50$0.03$0.08331M66K—
orcarouter$0.30$2.50$0.03—1M66K—
replicate · replicate$0.30$2.50—————
sap-ai-core$0.30$2.50$0.03—1M66K—
standardcompute$0.30$2.50$0.03$0.08331M50K—
vercel$0.30$2.50$0.03—1M66K—
vercel_ai_gateway · vercel-ai-gateway$0.30$2.50$0.03—1M66K—
vertex_ai-language-models$0.30$2.50$0.03—1M66K2026-10-20
zenmux$0.30$2.50$0.03—1M66K—
anyapi————1M66K—
kenari————1M66K—
qiniu-ai————1M64K—

Price history

append-only observations · a listing writes a row only when its price moves

DateHostField$/M
2026-09-02fastrouterCache read$0.0375
2026-10-22fastrouterCache read$0.03
2026-09-02kiloCache read$0.03
2026-10-21kiloCache read$0.015
2026-09-02openrouterCache read$0.03
2026-09-06openrouterCache read$0.03
2026-10-10openrouterCache read$0.03
2026-09-02zenmuxCache read$0.07
2026-10-22zenmuxCache read$0.03
2026-09-02kiloCache write$0.0833
2026-10-21kiloCache write$0.0417
2026-09-02openrouterCache write$0.0833
2026-09-19openrouterCache write$0.0833
2026-10-10openrouterCache write$0.0833
2026-09-02jiekouInput$0.27
2026-10-22jiekouInput$0.285
2026-09-02kiloInput$0.30
2026-10-21kiloInput$0.15
2026-09-02openrouterInput$0.30
2026-10-10openrouterInput$0.30
2026-09-02replicate · replicateInput$2.50
2026-09-19replicate · replicateInput$0.30
2026-09-02aihubmixOutput$2.50
2026-10-22aihubmixOutput$2.499
2026-09-02jiekouOutput$2.25
2026-10-22jiekouOutput$2.375
2026-09-02kiloOutput$2.50
2026-10-21kiloOutput$1.25
2026-09-02openrouterOutput$2.50
2026-10-10openrouterOutput$2.50

More from Google

most-hosted first