Gemini 3.5 Flash Lite
Google's cost-efficient Gemini 3.5 Flash Lite model for high-volume multimodal reasoning, tool use, and structured-output workloads.
- Reasoning
- Vision
- Audio Input
- Video Input
- Native PDF input
- Tool Calling
- Structured Output
Added Jul 21, 2026
Pricing
Auto routing · per 1M tokens- Input
- $0.30
- Output
- $2.50
- Cache read
- $0.03
Specifications
- Context window
- 1M
- Max output
- 65.5K
Benchmarks
Benchmarks
Sourced from LMArena.
Arena Score
1455.2
Overall Rank
#78 / 426
Votes
34,983
Confidence Interval
1450.7 - 1459.7
Category Scores
Coding
#85 / 421
9,667 votes
1502.1
Math
#99 / 408
1,585 votes
1442.1
Longer Query
#81 / 404
16,608 votes
1462.8
Creative Writing
#70 / 424
7,250 votes
1436.6
Instruction Following
#84 / 426
12,583 votes
1443.8
Hard Prompts
#83 / 426
23,225 votes
1475.1
Additional Categories22
Japanese
#44 / 288
581 votes
1447.6
German
#49 / 321
648 votes
1469.4
Korean
#60 / 289
639 votes
1421.1
Russian
#64 / 390
3,980 votes
1463.5
Polish
#65 / 243
515 votes
1459.2
French
#70 / 305
1,079 votes
1477.5
Industry Entertainment And Sports And Media
#71 / 424
8,994 votes
1434.1
Multi Turn
#73 / 424
5,780 votes
1467.2
Non English
#73 / 426
20,698 votes
1445.3
Industry Life And Physical And Social Science
#74 / 424
5,693 votes
1477.9
Spanish
#74 / 307
938 votes
1454.7
Exclude Ties
#75 / 426
26,364 votes
1459.5
Industry Writing And Literature And Language
#76 / 425
9,505 votes
1443.3
Industry Business And Management And Financial Operations
#79 / 419
6,501 votes
1453.3
Industry Medicine And Healthcare
#82 / 394
2,548 votes
1469.5
Industry Software And It Services
#85 / 426
14,039 votes
1488.9
English
#86 / 426
14,282 votes
1461.2
Hard Prompts English
#87 / 424
9,447 votes
1476.6
Chinese
#89 / 399
2,630 votes
1486.1
Industry Legal And Government
#94 / 397
2,964 votes
1455.8
Expert
#98 / 374
3,964 votes
1469.1
Industry Mathematical
#106 / 405
1,958 votes
1444.0
Published 2026-11-19 · Matched as Gemini 3.5 Flash-Lite · CC BY 4.0
LMArena DatasetLeaderboard results
| Benchmark | Score | Rank on NanoGPT |
|---|---|---|
Epoch AI Benchmarking Hub Epoch AI data, CC BY 4.0 · Updated Nov 21, 2026 | ||
Epoch Capabilities Index | 145.6 | #81 of 114 |
FrontierMath v2 (Tiers 1–3) | 26.0% | #71 of 84 |
GPQA Diamond | 83.3% | #74 of 119 |
FrontierMath v2 (Tier 4) | 0.0% | #68 of 68 |
OTIS Mock AIME 2024–2025 | 71.1% | #90 of 121 |
Chess Puzzles | 21.0% | #61 of 83 |
Mystery Game Puzzles | 19.0% | #53 of 80 |
Creative Writing v3 | 1559 | #51 of 91 |
Filtered and ranked to models on NanoGPT.
Providers
Provider information for this model’s automatic routing. These routes cannot be selected individually.
Choose or restrict providers with the API. Provider-selection prices include the selection markup. Provider selection guide · Pricing API (JSON, USD per 1K tokens)
| Cache write /M | ||||||
|---|---|---|---|---|---|---|
Recommended | $0.30 | $2.50 | $0.03 | $0.0833 | N/A | N/A |
| $0.30 | $2.50 | $0.03 | $0.0833 | N/A | N/A | |
| $0.30 | $2.50 | $0.03 | $0.0833 | N/A | N/A |