Gemini 3.5 Flash
- Context
- 1.05M
- 66K max output
- Input price
- $1.50
- per 1M tokens · Gemini API
- Output price
- $9.00
- per 1M tokens
- Epoch Capabilities Index
- 154.6
- #29 of 71 listed
- LMArena Text
- 1,477
- #21 of 79 listed
Where to run it
1 offering · prices as listed| Meter | Tier | Price |
|---|---|---|
| Input | Standard | $1.50 / 1M tokens |
| Batch | $0.75 / 1M tokens | |
| Flex tier | $0.75 / 1M tokens | |
| Priority tier | $2.70 / 1M tokens | |
| Output | Standard | $9.00 / 1M tokens |
| Batch | $4.50 / 1M tokens | |
| Flex tier | $4.50 / 1M tokens | |
| Priority tier | $16.20 / 1M tokens | |
| Cache read | Standard | $0.15 / 1M tokens |
| Batch | $0.075 / 1M tokens | |
| Flex tier | $0.080 / 1M tokens | |
| Priority tier | $0.27 / 1M tokens | |
Vendor-reported scores
Published by Google, under the conditions it states| Benchmark | Setting | Score |
|---|---|---|
| Terminal-Bench 2.1 | Terminus-2 harness, pass@1 | 76.2% |
| OSWorld-Verified | average of 5 runs, single attempt, max 100 steps | 78.4% |
| CharXiv Reasoning | no tools | 84.2% |
| MMMU-Pro | no tools, average of Standard (10 options) and Vision | 83.6% |
| Humanity's Last Exam | full set, text + MM | 40.2% |
Cited scores
From third-party leaderboards, not measured by us- Epoch Capabilities Index 154.6
#29 of 71 listed models · CI 152.6–157.2
Epoch AI, 'Capabilities & benchmarking'. Published online at epoch.ai. Retrieved from 'https://epoch.ai/benchmarks'. Licensed CC BY 4.0. · License · as of Sep 26, 2026 · changes we made
- LMArena Text 1,477
#21 of 79 listed models · high effort · CI 1,473–1,481 · 46,190 votes
Also listed: medium effort 1,474
LMArena (Arena), leaderboard-dataset on Hugging Face, licensed CC BY 4.0 · License · as of Sep 25, 2026 · changes we made
- Epoch AI · FrontierMath Tiers 1–3 62.8%
#27 of 56 listed models · high effort · ±2.87
Epoch AI, 'Capabilities & benchmarking'. Published online at epoch.ai. Retrieved from 'https://epoch.ai/benchmarks'. Licensed CC BY 4.0. · License · as of Sep 26, 2026 · changes we made
- Epoch AI · GPQA Diamond 92.8%
#14 of 64 listed models · high effort · ±1.64
Also listed: low effort 88.9% · minimal effort 86.4%
Epoch AI, 'Capabilities & benchmarking'. Published online at epoch.ai. Retrieved from 'https://epoch.ai/benchmarks'. Licensed CC BY 4.0. · License · as of Sep 26, 2026 · changes we made
- Epoch AI · SWE-bench Verified 79.3%
#2 of 23 listed models · high effort · ±1.84
Epoch AI, 'Capabilities & benchmarking'. Published online at epoch.ai. Retrieved from 'https://epoch.ai/benchmarks'. Licensed CC BY 4.0. · License · as of Sep 26, 2026 · changes we made
- LMArena Text · Chinese 1,514
#29 of 79 listed models · medium effort · CI 1,503–1,525 · 3,328 votes
Also listed: high effort 1,514
LMArena (Arena), leaderboard-dataset on Hugging Face, licensed CC BY 4.0 · License · as of Sep 25, 2026 · changes we made
- LMArena Text · Coding 1,507
#37 of 79 listed models · high effort · CI 1,501–1,514 · 13,289 votes
Also listed: medium effort 1,506
LMArena (Arena), leaderboard-dataset on Hugging Face, licensed CC BY 4.0 · License · as of Sep 25, 2026 · changes we made
- LMArena Vision 1,310
#8 of 49 listed models · high effort · CI 1,302–1,318 · 9,579 votes
Also listed: medium effort 1,307
LMArena (Arena), leaderboard-dataset on Hugging Face, licensed CC BY 4.0 · License · as of Sep 13, 2026 · changes we made