Noometry

Google, proprietary

Gemini 2.5 Flash-Lite

Gemini 2.5 Flash-Lite by Google ranks 211th of 354 ranked models on the Noometry Index as of October 2026, with a score of 37.0. Its strongest category is agentic & tool use, where it ranks 96th. API pricing starts at $0.10 per million input tokens and $0.40 per million output tokens, with a 1.05M-token context window.

Last verified

Noometry rank
#211 of 354
Index score
37.0
Evidence
Confirmed 33 results
Provider
Google
Released
June 17, 2025
Weights
Proprietary
Reasoning
Yes
Context window
1.05M
Max output
66K
Input price
$0.10 / M
Output price
$0.40 / M
Blended price
$0.18 / M
Output speed
172 tokens/s Kagi
Value
#40 of 219
Knowledge cutoff
January 2025
Input
text, image, audio, video, pdf

Category scores

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

  1. Coding 38.5
  2. Agentic & Tool Use 28.0
  3. Reasoning 22.2
  4. Math 38.0
  5. Knowledge 32.5
  6. Multimodal 29.1
  7. Multilingual 49.3
  8. Instruction Following 70.0
  9. Long Context 33.3
  10. Writing & Preference 56.8
CategoryScoreRankResults
Coding38.5#1732
Agentic & Tool Use28.0#961
Reasoning22.2#2054
Math38.0#1442
Knowledge32.5#2104
Multimodal29.1#1142
Multilingual49.3#1341
Instruction Following70.0#1682
Long Context33.3#2622
Writing & Preference56.8#1354

Strengths and weaknesses

Categories where Gemini 2.5 Flash-Lite places highest and lowest among the models ranked in each, with its score against that category's median.

Strongest categories

CategoryScorevs medianRank
Writing & Preference56.8+3.1#135 of 312, top 44%
Math38.0+1.4#144 of 327, top 45%
Multilingual49.3+1.9#134 of 297, top 46%

Weakest categories

CategoryScorevs medianRank
Multimodal29.1−9.4#114 of 128, top 90%
Long Context33.3−7.6#262 of 296, top 89%
Knowledge32.5−4.8#210 of 314, top 67%

Closest competitors

The models ranked just above and below Gemini 2.5 Flash-Lite. When scores are this close, price and speed are often the better way to choose.

ModelRankScoreBlended $/MSpeed
Step 3.7 Flash#20737.3$0.42—Compare
GPT-4.5#20837.2——Compare
Yi-Lightning#20937.1——Compare
Qwen Plus#21037.1$0.6037Compare
o3-mini#21236.7$1.93—Compare
Llama 3.1 Nemotron Ultra 253b v1#21336.7——Compare
Granite 4.0 H Small#21436.5——Compare
Command A#21536.5$4.3828Compare

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
WeirdML35.2%Epoch AI
WeirdML35.2%#91 of 119, top 77%16KEpoch AI
LMArena Coding1372no-thinkingLMArena2026-10-08
LMArena Coding1373#155 of 294, top 53%thinkingLMArena2026-10-08
ALE-Bench325.9#97 of 105, top 93%Epoch AI

Agentic & Tool Use

BenchmarkScorePositionSettingSourceDate
Berkeley Function Calling Leaderboard36.9%#28 of 49, top 58%fcBerkeley Function Calling Leaderboard

Reasoning

BenchmarkScorePositionSettingSourceDate
Kagi LLM Benchmark40.5%#76 of 99, top 77%Kagi LLM Benchmark
LMArena Hard Prompts1377#140 of 297, top 48%no-thinkingLMArena2026-10-08
LMArena Hard Prompts1374thinkingLMArena2026-10-08
DTBench62.8%#107 of 151, top 71%Epoch AI
LMCA18.1%#96 of 125, top 77%Epoch AI
Epoch Capabilities Index133.94#129 of 213, top 61%Epoch AI2025-06-17

Math

BenchmarkScorePositionSettingSourceDate
Omni-MATH48%#17 of 57, top 30%HELM Capabilities
LMArena Math1373#147 of 285, top 52%no-thinkingLMArena2026-10-08
LMArena Math1363thinkingLMArena2026-10-08

Knowledge

BenchmarkScorePositionSettingSourceDate
MMLU-Pro53.7%#50 of 58, top 87%HELM Capabilities
Vectara Hallucination Rate (lower is better)3.3%#2 of 96, top 3%Vectara Hallucination Leaderboard
GPQA (HELM)30.9%#52 of 57, top 92%HELM Capabilities
LMArena Expert1373#139 of 273, top 51%no-thinkingLMArena2026-10-08
LMArena Expert1366thinkingLMArena2026-10-08

Multimodal

BenchmarkScorePositionSettingSourceDate
LMArena Vision1198#79 of 122, top 65%no-thinkingLMArena2026-10-09
LMArena Vision1187thinkingLMArena2026-10-09
VPCT30%#24 of 24, top 100%Epoch AI

Multilingual

BenchmarkScorePositionSettingSourceDate
LMArena Non-English1369#134 of 297, top 46%no-thinkingLMArena2026-10-08
LMArena Non-English1360thinkingLMArena2026-10-08
LMArena Chinese1404#137 of 285, top 49%no-thinkingLMArena2026-10-08
LMArena Chinese1400thinkingLMArena2026-10-08
LMArena French1388#127 of 223, top 57%no-thinkingLMArena2026-10-08
LMArena French1387thinkingLMArena2026-10-08
LMArena German1389#105 of 231, top 46%no-thinkingLMArena2026-10-08
LMArena German1373thinkingLMArena2026-10-08
LMArena Japanese1359#92 of 211, top 44%no-thinkingLMArena2026-10-08
LMArena Japanese1350thinkingLMArena2026-10-08
LMArena Korean1350no-thinkingLMArena2026-10-08
LMArena Korean1360#95 of 213, top 45%thinkingLMArena2026-10-08
LMArena Russian1373#132 of 283, top 47%no-thinkingLMArena2026-10-08
LMArena Russian1364thinkingLMArena2026-10-08
LMArena Spanish1396#115 of 226, top 51%no-thinkingLMArena2026-10-08
LMArena Spanish1365thinkingLMArena2026-10-08

Instruction Following

BenchmarkScorePositionSettingSourceDate
IFEval81%#39 of 57, top 69%HELM Capabilities
LMArena Instruction Following1355no-thinkingLMArena2026-10-08
LMArena Instruction Following1367#134 of 298, top 45%thinkingLMArena2026-10-08

Long Context

BenchmarkScorePositionSettingSourceDate
Fiction.LiveBench47.2%#36 of 47, top 77%Epoch AI
LMArena Longer Query1373no-thinkingLMArena2026-10-08
LMArena Longer Query1373#136 of 291, top 47%thinkingLMArena2026-10-08

Writing & Preference

BenchmarkScorePositionSettingSourceDate
LMArena Text1379#136 of 297, top 46%no-thinkingLMArena2026-10-08
LMArena Text1369thinkingLMArena2026-10-08
LMArena Creative Writing1358no-thinkingLMArena2026-10-08
LMArena Creative Writing1367#113 of 295, top 39%thinkingLMArena2026-10-08
WildBench81.8%#21 of 57, top 37%HELM Capabilities
LMArena Multi-Turn1366#142 of 295, top 49%no-thinkingLMArena2026-10-08
LMArena Multi-Turn1360thinkingLMArena2026-10-08

API pricing by provider

RouteInput $/MOutput $/MCached input $/MChecked
google$0.10$0.40$0.012026-10-10
openrouter$0.10$0.40$0.012026-10-10
vertex$0.10$0.40$0.012026-10-10

All Google API prices → Estimate your cost →

Compare Gemini 2.5 Flash-Lite

Other Google models

Frequently asked questions

How good is Gemini 2.5 Flash-Lite?

Gemini 2.5 Flash-Lite by Google ranks 211th of 354 ranked models on the Noometry Index as of October 2026, with a score of 37.0. Its strongest category is agentic & tool use, where it ranks 96th. API pricing starts at $0.10 per million input tokens and $0.40 per million output tokens, with a 1.05M-token context window.

How much does Gemini 2.5 Flash-Lite cost?

Gemini 2.5 Flash-Lite costs $0.10 per million input tokens and $0.40 per million output tokens on Google's own API, with cached input at $0.01.

What is Gemini 2.5 Flash-Lite's context window?

Gemini 2.5 Flash-Lite accepts up to 1.05M tokens of input and can write up to 66K tokens in one response.

Is Gemini 2.5 Flash-Lite open source?

No. Gemini 2.5 Flash-Lite is proprietary and available only through Google's API and partner platforms.

How fast is Gemini 2.5 Flash-Lite?

Gemini 2.5 Flash-Lite generated about 172 output tokens per second in the Kagi LLM Benchmark's timed runs. Speed varies by provider, load and reasoning effort.

What are Gemini 2.5 Flash-Lite's strengths and weaknesses?

Relative to other ranked models, Gemini 2.5 Flash-Lite places best in writing & preference, math, multilingual and lowest in multimodal, long context, knowledge.

What is Gemini 2.5 Flash-Lite best at?

Its best category is agentic & tool use, where it ranks 96th on Noometry.