Gemini 3.8 Flash
Google's fast multimodal model for agentic workloads, including coding, tool use, image understanding, PDF and document extraction, audio, and video. Its capabilities, limits, reasoning behavior, and pricing currently mirror Gemini 3.7 Flash.
- Reasoning
- Vision
- Audio Input
- Video Input
- Native PDF input
- Tool Calling
- Structured Output
Added Sep 2, 2026
Pricing
Auto routing · per 1M tokens- Input
- $0.75
- Output
- $3.75
- Cache read
- $0.075
Specifications
- Context window
- 1M
- Max output
- 65.5K
Benchmarks
Benchmarks
Sourced from LMArena.
Arena Score
1497.3
Overall Rank
#12 / 426
Votes
73,351
Confidence Interval
1493.2 - 1501.3
Category Scores
Coding
#25 / 421
18,385 votes
1529.6
Math
#6 / 408
3,702 votes
1519.1
Longer Query
#11 / 404
33,487 votes
1510.4
Creative Writing
#11 / 424
16,669 votes
1488.8
Instruction Following
#16 / 426
26,673 votes
1492.5
Hard Prompts
#13 / 426
47,344 votes
1517.7
Additional Categories22
Japanese
#6 / 288
1,155 votes
1516.0
Russian
#6 / 390
10,667 votes
1515.4
Polish
#8 / 243
1,078 votes
1506.5
Industry Entertainment And Sports And Media
#11 / 424
19,530 votes
1478.5
Industry Writing And Literature And Language
#11 / 425
21,554 votes
1491.3
Multi Turn
#11 / 424
11,047 votes
1503.6
Non English
#11 / 426
45,702 votes
1487.9
Chinese
#12 / 399
6,102 votes
1547.4
German
#12 / 321
1,074 votes
1501.2
Industry Life And Physical And Social Science
#12 / 424
12,007 votes
1517.2
Exclude Ties
#13 / 426
54,879 votes
1514.7
English
#14 / 426
27,635 votes
1498.3
Industry Mathematical
#15 / 405
4,293 votes
1511.9
Korean
#15 / 289
1,072 votes
1461.2
Hard Prompts English
#17 / 424
17,984 votes
1515.0
Industry Medicine And Healthcare
#19 / 394
5,083 votes
1506.4
Expert
#20 / 374
7,851 votes
1526.3
Industry Business And Management And Financial Operations
#21 / 419
12,892 votes
1485.4
Spanish
#21 / 307
2,589 votes
1478.8
Industry Software And It Services
#26 / 426
26,800 votes
1520.2
French
#27 / 305
2,544 votes
1501.8
Industry Legal And Government
#30 / 397
6,096 votes
1495.5
Published 2026-11-19 · Matched as Gemini 3.8 Flash (High) · 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 22, 2026 | ||
Epoch Capabilities Index | 158.8 | #23 of 114 |
FrontierMath v2 (Tiers 1–3) | 68.4% | #33 of 84 |
GPQA Diamond | 95.4% | #8 of 119 |
SimpleQA Verified | 69.7% | #13 of 84 |
FrontierMath v2 (Tier 4) | 22.0% | #49 of 68 |
OTIS Mock AIME 2024–2025 | 98.9% | #26 of 121 |
MirrorCode | 38.9% | #28 of 41 |
Chess Puzzles | 58.0% | #14 of 83 |
Mystery Game Puzzles | 47.0% | #20 of 80 |
Creative Writing v3 | 1744 | #30 of 91 |
Longform Writing | 76.8 | #26 of 93 |
UGI Leaderboard UGI Leaderboard data, Apache-2.0 · Updated Nov 22, 2026 | ||
UGI Score | 46.7 | #51 of 125 |
UGI Willingness | 2.2 | #88 of 125 |
UGI Writing | 78.5 | #2 of 125 |
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 | $0.75 | $3.75 | $0.075 | $0.0417 | N/A | N/A |
| $0.75 | $3.75 | $0.075 | $0.0417 | N/A | N/A | |
| $0.75 | $3.75 | $0.075 | $0.0417 | N/A | N/A |