Google: Gemini 3.5 Flash Lite

Gemini 3.5 Flash Lite is a high-efficiency model from Google with upgraded agentic capabilities. It is suited for subagents that execute focused tasks within complex, multi-agent workflows.

google/gemini-3.5-flash-lite-20260721

Gemini 3.5 Flash Lite is a high-efficiency model from Google with upgraded agentic capabilities. It is suited for subagents that execute focused tasks within complex, multi-agent workflows.

Specifications & capabilities

Input
text, image, video, file, audio
Output
text
Context window · tokens
1048576
Knowledge cutoff
2026-03
Published model ID
gemini-3.5-flash-lite@default
Capabilities
audioInput, batch, computerUse, fileInput, multimodal, promptCaching, reasoning, structuredOutputs, toolCalling, videoInput, visionInput, webSearch

Evaluations

1–20 / 112

OpenRouter Benchmarks

73.333%

source id
openRouterBenchmarks
leaderboard name
tau2-bench Verified Airline · Accuracy · Google AI Studio
dimension
agentic
metric
tau2-bench Verified Airline · Accuracy
rank
3
sample size
100
aggregation scope
status
published
harness
url
https://github.com/OpenRouterTeam/benchmark-harness
name
OpenRouter benchmark harness
direction
higherIsBetter
run_count
2
conditions
window
newest 90 days; successful visible nonzero runs
dataset
benchmark-leaderboard
benchmark
id
tau_bench_verified_airline
name
tau2-bench Verified Airline
record_kind
aggregate
provider_name
Google AI Studio
attribution
Source: OpenRouter (openrouter.ai), licensed under CC BY 4.0: reuse and republish with attribution to OpenRouter.
last_run_at
2026-10-01T19:05:46.177Z
license_url
https://creativecommons.org/licenses/by/4.0/
methodology
Rows aggregate the benchmark-runs dataset exactly as the leaderboard pages do, over each benchmark's non-zero-score runs from the newest 90 days: accuracy is the task-weighted mean of per-run accuracy, accuracy_std_dev is the population standard deviation of per-run accuracy (null with a single run), and the per-task averages divide summed cost, generation time, and output tokens by summed tasks. Rows are ordered by accuracy descending, then model name.
metric_kind
tau2-bench Verified Airline · Accuracy
rank_method
competition rank among reported evaluations in identical conditions, including unlinked observations
sample_size
100
source_file
benchmark-leaderboard.json
source_page
https://openrouter.ai/benchmarks/tau2-bench-airline
measurements
accuracy
0.7333330000000001
accuracy_std_dev
0
cost_per_task_usd
0.10928178
output_tokens_per_task
18824.05
latency_seconds_per_task
75.47944
published_at
2026-11-16T01:46:02.653Z
snapshot_date
2026-11-16
source_revision
7ccaf17664252a3053dcb7df8ad0a46e72954a3b4524ddc1c8a7ef8a41b15568
transformations
TensorTrace preserves source results, converts latency to seconds and derives ranks within matching conditions. Individual runs have no derived rank.
metric_population
10
benchmark_evaluation
true
openrouter_permaslug
google/gemini-3.5-flash-lite-20260721
source model id
google/gemini-3.5-flash-lite-20260721
Original result ↗

OpenRouter Benchmarks

78.333%

source id
openRouterBenchmarks
leaderboard name
tau2-bench Verified Airline · Accuracy · auto-routing
dimension
agentic
metric
tau2-bench Verified Airline · Accuracy
rank
6
sample size
100
aggregation scope
status
published
harness
url
https://github.com/OpenRouterTeam/benchmark-harness
name
OpenRouter benchmark harness
direction
higherIsBetter
run_count
2
conditions
window
newest 90 days; successful visible nonzero runs
dataset
benchmark-leaderboard
benchmark
id
tau_bench_verified_airline
name
tau2-bench Verified Airline
record_kind
aggregate
provider_name
auto-routing
attribution
Source: OpenRouter (openrouter.ai), licensed under CC BY 4.0: reuse and republish with attribution to OpenRouter.
last_run_at
2026-10-01T19:04:42.463Z
license_url
https://creativecommons.org/licenses/by/4.0/
methodology
Rows aggregate the benchmark-runs dataset exactly as the leaderboard pages do, over each benchmark's non-zero-score runs from the newest 90 days: accuracy is the task-weighted mean of per-run accuracy, accuracy_std_dev is the population standard deviation of per-run accuracy (null with a single run), and the per-task averages divide summed cost, generation time, and output tokens by summed tasks. Rows are ordered by accuracy descending, then model name.
metric_kind
tau2-bench Verified Airline · Accuracy
rank_method
competition rank among reported evaluations in identical conditions, including unlinked observations
sample_size
100
source_file
benchmark-leaderboard.json
source_page
https://openrouter.ai/benchmarks/tau2-bench-airline
measurements
accuracy
0.7833330000000001
accuracy_std_dev
0.009999999999999953
cost_per_task_usd
0.07443635
output_tokens_per_task
21518.67
latency_seconds_per_task
98.39715
published_at
2026-11-16T01:46:02.653Z
snapshot_date
2026-11-16
source_revision
7ccaf17664252a3053dcb7df8ad0a46e72954a3b4524ddc1c8a7ef8a41b15568
transformations
TensorTrace preserves source results, converts latency to seconds and derives ranks within matching conditions. Individual runs have no derived rank.
metric_population
128
benchmark_evaluation
true
openrouter_permaslug
google/gemini-3.5-flash-lite-20260721
source model id
google/gemini-3.5-flash-lite-20260721
Original result ↗

OpenRouter Benchmarks

84.091%

source id
openRouterBenchmarks
leaderboard name
GPQA Diamond · Accuracy · Google AI Studio
dimension
knowledge
metric
GPQA Diamond · Accuracy
rank
7
sample size
396
aggregation scope
status
published
harness
url
https://github.com/OpenRouterTeam/benchmark-harness
name
OpenRouter benchmark harness
direction
higherIsBetter
run_count
2
conditions
window
newest 90 days; successful visible nonzero runs
dataset
benchmark-leaderboard
benchmark
id
gpqa_diamond
name
GPQA Diamond
record_kind
aggregate
provider_name
Google AI Studio
attribution
Source: OpenRouter (openrouter.ai), licensed under CC BY 4.0: reuse and republish with attribution to OpenRouter.
last_run_at
2026-10-01T19:13:29.958Z
license_url
https://creativecommons.org/licenses/by/4.0/
methodology
Rows aggregate the benchmark-runs dataset exactly as the leaderboard pages do, over each benchmark's non-zero-score runs from the newest 90 days: accuracy is the task-weighted mean of per-run accuracy, accuracy_std_dev is the population standard deviation of per-run accuracy (null with a single run), and the per-task averages divide summed cost, generation time, and output tokens by summed tasks. Rows are ordered by accuracy descending, then model name.
metric_kind
GPQA Diamond · Accuracy
rank_method
competition rank among reported evaluations in identical conditions, including unlinked observations
sample_size
396
source_file
benchmark-leaderboard.json
source_page
https://openrouter.ai/benchmarks/gpqa-diamond
measurements
accuracy
0.8409089999999999
accuracy_std_dev
0.010943000000000036
cost_per_task_usd
0.03708839646464646
output_tokens_per_task
14746.045454545454
latency_seconds_per_task
47.51584343434344
published_at
2026-11-16T01:46:02.653Z
snapshot_date
2026-11-16
source_revision
7ccaf17664252a3053dcb7df8ad0a46e72954a3b4524ddc1c8a7ef8a41b15568
transformations
TensorTrace preserves source results, converts latency to seconds and derives ranks within matching conditions. Individual runs have no derived rank.
metric_population
12
benchmark_evaluation
true
openrouter_permaslug
google/gemini-3.5-flash-lite-20260721
source model id
google/gemini-3.5-flash-lite-20260721
Original result ↗

OpenRouter Benchmarks

76.095%

source id
openRouterBenchmarks
leaderboard name
tau2-bench Verified Airline · Accuracy · Google
dimension
agentic
metric
tau2-bench Verified Airline · Accuracy
rank
11
sample size
350
aggregation scope
status
published
harness
url
https://github.com/OpenRouterTeam/benchmark-harness
name
OpenRouter benchmark harness
direction
higherIsBetter
run_count
7
conditions
window
newest 90 days; successful visible nonzero runs
dataset
benchmark-leaderboard
benchmark
id
tau_bench_verified_airline
name
tau2-bench Verified Airline
record_kind
aggregate
provider_name
Google
attribution
Source: OpenRouter (openrouter.ai), licensed under CC BY 4.0: reuse and republish with attribution to OpenRouter.
last_run_at
2026-10-01T19:05:47.886Z
license_url
https://creativecommons.org/licenses/by/4.0/
methodology
Rows aggregate the benchmark-runs dataset exactly as the leaderboard pages do, over each benchmark's non-zero-score runs from the newest 90 days: accuracy is the task-weighted mean of per-run accuracy, accuracy_std_dev is the population standard deviation of per-run accuracy (null with a single run), and the per-task averages divide summed cost, generation time, and output tokens by summed tasks. Rows are ordered by accuracy descending, then model name.
metric_kind
tau2-bench Verified Airline · Accuracy
rank_method
competition rank among reported evaluations in identical conditions, including unlinked observations
sample_size
350
source_file
benchmark-leaderboard.json
source_page
https://openrouter.ai/benchmarks/tau2-bench-airline
measurements
accuracy
0.7609522857142857
accuracy_std_dev
0.01649574672206044
cost_per_task_usd
0.08085598000000001
output_tokens_per_task
22435.354285714286
latency_seconds_per_task
105.36985714285714
published_at
2026-11-16T01:46:02.653Z
snapshot_date
2026-11-16
source_revision
7ccaf17664252a3053dcb7df8ad0a46e72954a3b4524ddc1c8a7ef8a41b15568
transformations
TensorTrace preserves source results, converts latency to seconds and derives ranks within matching conditions. Individual runs have no derived rank.
metric_population
36
benchmark_evaluation
true
openrouter_permaslug
google/gemini-3.5-flash-lite-20260721
source model id
google/gemini-3.5-flash-lite-20260721
Original result ↗

OpenRouter Benchmarks

83.983%

source id
openRouterBenchmarks
leaderboard name
GPQA Diamond · Accuracy · Google
dimension
knowledge
metric
GPQA Diamond · Accuracy
rank
21
sample size
1386
aggregation scope
status
published
harness
url
https://github.com/OpenRouterTeam/benchmark-harness
name
OpenRouter benchmark harness
direction
higherIsBetter
run_count
7
conditions
window
newest 90 days; successful visible nonzero runs
dataset
benchmark-leaderboard
benchmark
id
gpqa_diamond
name
GPQA Diamond
record_kind
aggregate
provider_name
Google
attribution
Source: OpenRouter (openrouter.ai), licensed under CC BY 4.0: reuse and republish with attribution to OpenRouter.
last_run_at
2026-10-01T19:13:57.987Z
license_url
https://creativecommons.org/licenses/by/4.0/
methodology
Rows aggregate the benchmark-runs dataset exactly as the leaderboard pages do, over each benchmark's non-zero-score runs from the newest 90 days: accuracy is the task-weighted mean of per-run accuracy, accuracy_std_dev is the population standard deviation of per-run accuracy (null with a single run), and the per-task averages divide summed cost, generation time, and output tokens by summed tasks. Rows are ordered by accuracy descending, then model name.
metric_kind
GPQA Diamond · Accuracy
rank_method
competition rank among reported evaluations in identical conditions, including unlinked observations
sample_size
1386
source_file
benchmark-leaderboard.json
source_page
https://openrouter.ai/benchmarks/gpqa-diamond
measurements
accuracy
0.8398265714285714
accuracy_std_dev
0.008557108588054464
cost_per_task_usd
0.03983322727272727
output_tokens_per_task
14785.20634920635
latency_seconds_per_task
52.95570346320346
published_at
2026-11-16T01:46:02.653Z
snapshot_date
2026-11-16
source_revision
7ccaf17664252a3053dcb7df8ad0a46e72954a3b4524ddc1c8a7ef8a41b15568
transformations
TensorTrace preserves source results, converts latency to seconds and derives ranks within matching conditions. Individual runs have no derived rank.
metric_population
40
benchmark_evaluation
true
openrouter_permaslug
google/gemini-3.5-flash-lite-20260721
source model id
google/gemini-3.5-flash-lite-20260721
Original result ↗

LMArena

1,266.558

source id
lmarena
leaderboard name
LMArena Vision Style Control Humor
dimension
multimodal
metric
Arena rating
rank
26
sample size
484
evaluated at
2026-11-09T00:00:00.000Z
source model id
Gemini 3.5 Flash-Lite
confidence lower
1238.8830938067113
confidence upper
1294.2330542503535
Original result ↗

LMArena

1,274.07

source id
lmarena
leaderboard name
LMArena Vision Humor
dimension
multimodal
metric
Arena rating
rank
36
sample size
484
evaluated at
2026-11-09T00:00:00.000Z
source model id
Gemini 3.5 Flash-Lite
confidence lower
1246.45644480443
confidence upper
1301.684013862187
Original result ↗

LMArena

1,265.799

source id
lmarena
leaderboard name
LMArena Vision Style Control Overall
dimension
multimodal
metric
Arena rating
rank
38
sample size
10800
evaluated at
2026-11-09T00:00:00.000Z
source model id
Gemini 3.5 Flash-Lite
confidence lower
1258.315898088031
confidence upper
1273.2812961512757
Original result ↗

LMArena

1,279.806

source id
lmarena
leaderboard name
LMArena Vision Style Control Ocr
dimension
multimodal
metric
Arena rating
rank
39
sample size
7376
evaluated at
2026-11-09T00:00:00.000Z
source model id
Gemini 3.5 Flash-Lite
confidence lower
1271.8423994094667
confidence upper
1287.7688456966569
Original result ↗

LMArena

1,265.84

source id
lmarena
leaderboard name
LMArena Vision Style Control English
dimension
multimodal
metric
Arena rating
rank
39
sample size
3615
evaluated at
2026-11-09T00:00:00.000Z
source model id
Gemini 3.5 Flash-Lite
confidence lower
1254.7171941801514
confidence upper
1276.962232015616
Original result ↗

LMArena

1,252.193

source id
lmarena
leaderboard name
LMArena Vision Style Control Creative Writing Vision
dimension
multimodal
metric
Arena rating
rank
42
sample size
998
evaluated at
2026-11-09T00:00:00.000Z
source model id
Gemini 3.5 Flash-Lite
confidence lower
1232.3517777630188
confidence upper
1272.0339630117826
Original result ↗

LMArena

1,305.609

source id
lmarena
leaderboard name
LMArena Vision Style Control Chinese
dimension
multimodal
metric
Arena rating
rank
43
sample size
615
evaluated at
2026-11-09T00:00:00.000Z
source model id
Gemini 3.5 Flash-Lite
confidence lower
1280.7987879280477
confidence upper
1330.4184574523538
Original result ↗

LMArena

1,277.946

source id
lmarena
leaderboard name
LMArena Vision Ocr
dimension
multimodal
metric
Arena rating
rank
43
sample size
7376
evaluated at
2026-11-09T00:00:00.000Z
source model id
Gemini 3.5 Flash-Lite
confidence lower
1270.0090200007223
confidence upper
1285.8831345817619
Original result ↗

LMArena

1,267.983

source id
lmarena
leaderboard name
LMArena Vision Overall
dimension
multimodal
metric
Arena rating
rank
45
sample size
10800
evaluated at
2026-11-09T00:00:00.000Z
source model id
Gemini 3.5 Flash-Lite
confidence lower
1260.5262851905284
confidence upper
1275.439677752136
Original result ↗

LMArena

1,444.544

source id
lmarena
leaderboard name
LMArena Text Style Control Japanese
dimension
style
metric
Arena rating
rank
46
sample size
587
evaluated at
2026-11-09T00:00:00.000Z
source model id
Gemini 3.5 Flash-Lite
confidence lower
1418.0542181024
confidence upper
1471.0336464867912
Original result ↗

LMArena

1,288.645

source id
lmarena
leaderboard name
LMArena Vision Homework
dimension
multimodal
metric
Arena rating
rank
46
sample size
1274
evaluated at
2026-11-09T00:00:00.000Z
source model id
Gemini 3.5 Flash-Lite
confidence lower
1271.5949553968228
confidence upper
1305.6952452789162
Original result ↗

LMArena

1,468.895

source id
lmarena
leaderboard name
LMArena Text Style Control German
dimension
style
metric
Arena rating
rank
47
sample size
645
evaluated at
2026-11-09T00:00:00.000Z
source model id
Gemini 3.5 Flash-Lite
confidence lower
1445.6460175829257
confidence upper
1492.1449566096317
Original result ↗

LMArena

1,288.138

source id
lmarena
leaderboard name
LMArena Vision Style Control Homework
dimension
multimodal
metric
Arena rating
rank
47
sample size
1274
evaluated at
2026-11-09T00:00:00.000Z
source model id
Gemini 3.5 Flash-Lite
confidence lower
1271.0669301673472
confidence upper
1305.2087141767036
Original result ↗

LMArena

1,287.161

source id
lmarena
leaderboard name
LMArena Vision Style Control Diagram
dimension
multimodal
metric
Arena rating
rank
47
sample size
2708
evaluated at
2026-11-09T00:00:00.000Z
source model id
Gemini 3.5 Flash-Lite
confidence lower
1274.9873261536104
confidence upper
1299.3343489839153
Original result ↗

LMArena

1,426.042

source id
lmarena
leaderboard name
LMArena Text Japanese
dimension
general
metric
Arena rating
rank
49
sample size
587
evaluated at
2026-11-09T00:00:00.000Z
source model id
Gemini 3.5 Flash-Lite
confidence lower
1399.5639415037278
confidence upper
1452.5203744333332
Original result ↗

Published prices

61–61 / 61

google/gemini-3.5-flash-lite ↗

$0.01

Inference provider
vercelAIGatewayCatalog
Platform
vertex
Unit
request
Currency
USD
Billing conditions
charge
web_search
endpoint
vertex | google/gemini-3.5-flash-lite
model_type
language
billing_scope
via Vercel AI Gateway
Billing label
Web search
Pricing via
Vercel AI Gateway

Provider specifications & access

1–4 / 4

gemini-3.5-flash-lite ↗

source model id
gemini-3.5-flash-lite
hosting
gateway
platforms
  • Cheaper Inference
context length
1000000
max output tokens
65536
architecture
input_modalities
  • text
  • image
  • video
output_modalities
  • text
capabilities
  • reasoning
  • structuredOutputs
  • toolCalling
  • videoInput
  • visionInput
  • webSearch

Gemini 3.5 Flash-Lite ↗

source model id
google/gemini-3.5-flash-lite
hosting
third-party
platforms
  • Google Vertex AI
context length
1048576
max output tokens
65536
architecture
input_modalities
  • text
output_modalities
  • text
capabilities

    Gemini 3.5 Flash Lite ↗

    source model id
    google/gemini-3.5-flash-lite
    hosting
    vercel-gateway
    platforms
    • Vercel AI Gateway
    context length
    1000000
    max output tokens
    65000
    architecture
    input_modalities
    • text
    • image
    • pdf
    • video
    output_modalities
    • text
    capabilities
    • promptCaching
    • reasoning
    • structuredOutputs
    • toolCalling
    • videoInput
    • visionInput
    • webSearch
    regions
    • eu
    • us

    Gemini 3.5 Flash-Lite ↗

    source model id
    publishers/google/models/gemini-3.5-flash-lite@default
    hosting
    vertex-managed
    platforms
    • Google Cloud Vertex AI
    context length
    1048576
    max output tokens
    65536
    architecture
    input_modalities
    • audio
    • image
    • text
    • video
    output_modalities
    • text
    capabilities
    • batch
    • computerUse
    • multimodal
    • promptCaching
    • reasoning
    • structuredOutputs
    • toolCalling
    • visionInput

    Safety assessments

    1–1 / 1

    Foundation Model Transparency Index 2025 ↗

    41 / 100

    source id
    stanfordFMTI
    canonical model id
    google/gemini-3.5-flash-lite-20260721
    taxonomy
    transparency
    hazard category
    transparency
    hazard name
    Organization transparency
    policy area
    Google
    assessment type
    organization-level disclosure
    metric
    disclosure_score
    result label
    41 / 100
    sample size
    100
    updated at
    2026-11-09T18:08:05.227Z

    Source specification observations

    1–11 / 11

    cloudflareWorkersAI ↗

    field
    context_length
    value
    1048576
    normalized value
    1048576
    scope
    source-model
    source id
    cloudflareWorkersAI
    confidence
    1

    openRouter ↗

    field
    context_length
    value
    1048576
    normalized value
    1048576
    scope
    source-model
    source id
    openRouter
    confidence
    1

    openRouter ↗

    field
    context_length
    value
    1048576
    normalized value
    1048576
    scope
    source-model
    source id
    openRouter
    confidence
    1

    vercelAIGatewayCatalog ↗

    field
    context_length
    value
    1000000
    normalized value
    1000000
    scope
    source-model
    source id
    vercelAIGatewayCatalog
    confidence
    1

    vertexAI ↗

    field
    context_length
    value
    1048576
    normalized value
    1048576
    scope
    source-model
    source id
    vertexAI
    confidence
    1

    vercelAIGatewayCatalog ↗

    field
    knowledge_cutoff
    reported field
    knowledge_cutoff
    value
    2026-03
    normalized value
    2026-03
    precision
    month
    scope
    source-model
    source id
    vercelAIGatewayCatalog
    confidence
    1

    cloudflareWorkersAI ↗

    field
    max_output_tokens
    value
    65536
    normalized value
    65536
    scope
    source-model
    source id
    cloudflareWorkersAI
    confidence
    1

    openRouter ↗

    field
    max_output_tokens
    value
    65536
    normalized value
    65536
    scope
    source-model
    source id
    openRouter
    confidence
    1

    openRouter ↗

    field
    max_output_tokens
    value
    65536
    normalized value
    65536
    scope
    source-model
    source id
    openRouter
    confidence
    1

    vercelAIGatewayCatalog ↗

    field
    max_output_tokens
    value
    65000
    normalized value
    65000
    scope
    source-model
    source id
    vercelAIGatewayCatalog
    confidence
    1

    vertexAI ↗

    field
    max_output_tokens
    value
    65536
    normalized value
    65536
    scope
    source-model
    source id
    vertexAI
    confidence
    1

    Original sources

    1–16 / 16

    Google · Model reference ↗

    Google Cloud · Model page ↗

    OpenRouter · Model page ↗

    OpenRouter · Model page ↗

    Vercel AI Gateway · Model page ↗

    Google Cloud · Documentation ↗

    GitHub · Source snapshot ↗

    api.cheaperinference.com · Source snapshot ↗

    Cloudflare AI · Catalog ↗

    Epoch AI · Catalog ↗

    GitHub · Catalog ↗

    Google Cloud · Catalog ↗

    LMArena · Catalog ↗

    OpenRouter · Catalog ↗

    OpenRouter · Catalog ↗

    Vercel AI Gateway · Catalog ↗

    Source links

    1–10 / 10

    Explore more models

    Models

    Loading
    EVIDENCE / COMPARISON

    Benchmarks

    Explore every benchmark

    Loading

    These charts contain model-linked results. The charts above also include unlinked source profiles. Compare scores within each evaluation. Model quality, usage, transparency and system performance measure different things.

    Install TensorTrace