Noometry

Google, proprietary

Gemini 3.5 Flash

Gemini 3.5 Flash by Google ranks 32nd of 354 ranked models on the Noometry Index as of October 2026, with a score of 54.2. Its strongest category is knowledge, where it ranks 11th. API pricing starts at $1.50 per million input tokens and $9 per million output tokens, with a 1.05M-token context window.

Last verified

Noometry rank
#32 of 354
Index score
54.2
Evidence
Confirmed 54 results
Provider
Google
Released
May 19, 2026
Weights
Proprietary
Reasoning
Yes
Context window
1.05M
Max output
66K
Input price
$1.50 / M
Output price
$9 / M
Blended price
$3.38 / M
Output speed
Not measured
Value
#161 of 219
Knowledge cutoff
January 2025
Input
text, image, video, audio, pdf

Category scores

Each category score combines every public result we have in that category.

  1. Coding 49.4
  2. Agentic & Tool Use 24.7
  3. Reasoning 62.8
  4. Math 60.7
  5. Knowledge 66.3
  6. Multimodal 45.7
  7. Multilingual 57.0
  8. Instruction Following 77.0
  9. Long Context 45.4
  10. Writing & Preference 65.5
CategoryScoreRankResults
Coding49.4#496
Agentic & Tool Use24.7#1143
Reasoning62.8#1813
Math60.7#366
Knowledge66.3#113
Multimodal45.7#152
Multilingual57.0#131
Instruction Following77.0#301
Long Context45.4#381
Writing & Preference65.5#474

Strengths and weaknesses

Categories where Gemini 3.5 Flash places highest and lowest among the models ranked in each, with its score against that category's median.

Strongest categories

CategoryScorevs medianRank
Knowledge66.3+29.0#11 of 314, top 4%
Multilingual57.0+9.6#13 of 297, top 5%
Reasoning62.8+39.2#18 of 350, top 6%

Weakest categories

CategoryScorevs medianRank
Agentic & Tool Use24.7−5.6#114 of 154, top 75%
Writing & Preference65.5+11.7#47 of 312, top 16%
Coding49.4+10.7#49 of 340, top 15%

Closest competitors

The models ranked just above and below Gemini 3.5 Flash. When scores are this close, price and speed are often the better way to choose.

ModelRankScoreBlended $/MSpeed
Gemini 3 Pro#2854.8—1Compare
Claude Sonnet 5#2954.6$4—Compare
GPT-5.6 Luna#3054.6$0.4512Compare
DeepSeek V4 Pro#3154.3$0.9916Compare
Gemini 3.6 Flash#3354.1$1.50—Compare
GPT-5.2#3454.1$4.8115Compare
DeepSeek V4 Flash#3553.6$0.266Compare
GPT-6 Luna#3653.3$0.20—Compare

Sponsored placements are available on pages like this one. Advertise on Noometry

Benchmark results

Every published result we track, with its source. Bold rows are the ones used for ranking; where several exist we prefer independent runs over self-reported numbers.

Coding

BenchmarkScorePositionSettingSourceDate
SWE-bench Verified79.3%#3 of 32, top 10%highEpoch AI2026-06-01
DeepSWE36.1%highEpoch AI
DeepSWE37.4%#26 of 29, top 90%mediumEpoch AI
LMArena WebDev1499#48 of 113, top 43%highLMArena2026-10-08
SciCode53.1%#34 of 121, top 29%highEpoch AI
WeirdML62.6%#26 of 119, top 22%highEpoch AI
LMArena Coding1492#28 of 294, top 10%highLMArena2026-10-08
ALE-Bench911.02#49 of 105, top 47%highEpoch AI

Agentic & Tool Use

BenchmarkScorePositionSettingSourceDate
APEX-Agents27.5%#45 of 49, top 92%Epoch AI
GBAEval6.7%#14 of 23, top 61%Epoch AI
GDP.pdf14%#28 of 36, top 78%Epoch AI
GDP.pdf14%mediumEpoch AI
Vending-Bench 25,396#26 of 60, top 44%Epoch AI

Reasoning

BenchmarkScorePositionSettingSourceDate
ARC-AGI-272.1%#20 of 83, top 25%highEpoch AI
ARC-AGI-28.9%minimalEpoch AI
SimpleBench76.7%#5 of 77, top 7%Epoch AI
NYT Connections (extended)92.6%#12 of 91, top 14%Lech Mazur benchmarks
ARC-AGI-192.5%#22 of 83, top 27%highEpoch AI
ARC-AGI-148.8%minimalEpoch AI
CritPt13.1%#41 of 134, top 31%highEpoch AI
Chess Puzzles50%#10 of 129, top 8%highEpoch AI2026-05-28
Chess Puzzles45%lowEpoch AI2026-08-06
Chess Puzzles43%minimalEpoch AI2026-07-15
EnigmaEval25.4%#4 of 38, top 11%highEpoch AI
EBR-Bench4.8%#22 of 24, top 92%highEpoch AI2026-06-25
LMArena Hard Prompts1488#18 of 297, top 7%highLMArena2026-10-08
Mystery Game Puzzles32%#24 of 74, top 33%highEpoch AI2026-07-27
Mystery Game Puzzles28%lowEpoch AI2026-08-05
DTBench94.7%#19 of 151, top 13%highEpoch AI
LMCA47.1%#27 of 125, top 22%highEpoch AI
Surface Evolver Bench58.1%#11 of 25, top 44%mediumEpoch AI
Epoch Capabilities Index154.46#35 of 213, top 17%Epoch AI2026-05-19
ForecastBench59#44 of 72, top 62%Epoch AI

Math

BenchmarkScorePositionSettingSourceDate
FrontierMath (Tiers 1-3)62.8%#34 of 81, top 42%highEpoch AI2026-06-10
FrontierMath Tier 426.8%#36 of 63, top 58%highEpoch AI2026-06-10
MathArena Final-Answer Competitions76.3%#9 of 29, top 32%MathArena
OTIS Mock AIME 2024-202595.6%#32 of 173, top 19%highEpoch AI2026-05-25
OTIS Mock AIME 2024-202588.9%lowEpoch AI2026-08-06
OTIS Mock AIME 2024-202580%minimalEpoch AI2026-07-15
ProofBench31%#41 of 77, top 54%highEpoch AI
LMArena Math1504#11 of 285, top 4%highLMArena2026-10-08
FrontierMath (Feb 2025 set)39%#10 of 68, top 15%highEpoch AI2026-05-22
FrontierMath Tier 4 (v1)14.6%#15 of 55, top 28%highEpoch AI2026-05-25

Knowledge

BenchmarkScorePositionSettingSourceDate
GPQA Diamond92.8%#19 of 186, top 11%highEpoch AI2026-05-22
GPQA Diamond88.9%lowEpoch AI2026-08-06
GPQA Diamond86.4%minimalEpoch AI2026-07-15
SimpleQA Verified66.2%#11 of 77, top 15%highEpoch AI2026-08-27
LMArena Expert1495#27 of 273, top 10%highLMArena2026-10-08

Multimodal

BenchmarkScorePositionSettingSourceDate
LMArena Vision1310#10 of 122, top 9%highLMArena2026-10-09
Blueprint-Bench 233.6%#9 of 31, top 30%Epoch AI
LMArena Document1463#16 of 38, top 43%mediumLMArena2026-09-13

Multilingual

BenchmarkScorePositionSettingSourceDate
LMArena Non-English1476#13 of 297, top 5%highLMArena2026-10-08
LMArena Chinese1526#22 of 285, top 8%mediumLMArena2026-10-08
LMArena French1490#20 of 223, top 9%highLMArena2026-10-08
LMArena German1492#12 of 231, top 6%mediumLMArena2026-10-08
LMArena Japanese1486#12 of 211, top 6%highLMArena2026-10-08
LMArena Korean1451#17 of 213, top 8%mediumLMArena2026-10-08
LMArena Russian1493#11 of 283, top 4%highLMArena2026-10-08
LMArena Spanish1480#13 of 226, top 6%highLMArena2026-10-08

Instruction Following

BenchmarkScorePositionSettingSourceDate
LMArena Instruction Following1467#26 of 298, top 9%highLMArena2026-10-08

Long Context

BenchmarkScorePositionSettingSourceDate
LMArena Longer Query1482#20 of 291, top 7%highLMArena2026-10-08

Writing & Preference

BenchmarkScorePositionSettingSourceDate
LMArena Text1482#14 of 297, top 5%highLMArena2026-10-08
LMArena Creative Writing1470#13 of 295, top 5%mediumLMArena2026-10-08
EQ-Bench 41087#24 of 28, top 86%EQ-Bench
LMArena Multi-Turn1481#19 of 295, top 7%highLMArena2026-10-08

API pricing by provider

RouteInput $/MOutput $/MCached input $/MChecked
google$1.50$9$0.152026-10-10
openrouter$1.50$9$0.152026-10-10
vertex$1.50$9$0.152026-10-10

All Google API prices → Estimate your cost →

Compare Gemini 3.5 Flash

Other Google models

Frequently asked questions

How good is Gemini 3.5 Flash?

Gemini 3.5 Flash by Google ranks 32nd of 354 ranked models on the Noometry Index as of October 2026, with a score of 54.2. Its strongest category is knowledge, where it ranks 11th. API pricing starts at $1.50 per million input tokens and $9 per million output tokens, with a 1.05M-token context window.

How much does Gemini 3.5 Flash cost?

Gemini 3.5 Flash costs $1.50 per million input tokens and $9 per million output tokens on Google's own API, with cached input at $0.15.

What is Gemini 3.5 Flash's context window?

Gemini 3.5 Flash accepts up to 1.05M tokens of input and can write up to 66K tokens in one response.

Is Gemini 3.5 Flash open source?

No. Gemini 3.5 Flash is proprietary and available only through Google's API and partner platforms.

What are Gemini 3.5 Flash's strengths and weaknesses?

Relative to other ranked models, Gemini 3.5 Flash places best in knowledge, multilingual, reasoning and lowest in agentic & tool use, writing & preference, coding.

What is Gemini 3.5 Flash best at?

Its best category is knowledge, where it ranks 11th on Noometry.