# Gemini 3.8 Flash vs o3-mini

> Gemini 3.8 Flash is the stronger model overall, scoring 59.5 to 36.7 on the Noometry Index.

- Canonical page: https://noometry.com/compare/gemini-3-8-flash-vs-o3-mini
- Last updated: 2026-10-10
- Shared benchmarks: 29

## Summary

- They share 29 benchmarks with published results for both. Gemini 3.8 Flash scores higher in 9 categories and o3-mini in 0 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where Gemini 3.8 Flash leads 64.7 to 16.3.
- The biggest single-benchmark swing is SimpleQA Verified: 69.7% for Gemini 3.8 Flash and 15.3% for o3-mini.
- Gemini 3.8 Flash is cheaper at $0.75 / $3.75 per million input/output tokens, against $1.10 / $4.40 for o3-mini.
- Gemini 3.8 Flash accepts more context: 1.05M tokens versus 200K.

## Snapshot

| | Gemini 3.8 Flash | o3-mini |
|---|---|---|
| Provider | Google | OpenAI |
| Noometry Index | 59.5 | 36.7 |
| Rank | 13 | 205 |
| Context | 1.05M | 200K |
| Input $/M | $0.75 | $1.10 |
| Output $/M | $3.75 | $4.40 |
| Weights | Proprietary | Proprietary |

## Coding

- Gemini 3.8 Flash: 56.4 (#17)
- o3-mini: 41.4 (#116)

| Benchmark | Gemini 3.8 Flash | o3-mini |
|---|---|---|
| SciCode | 56.6% | 39.8% |
| LMArena Coding | 1512 | 1378 |
| DeepSWE | 73.8% | — |
| FrontierCode | 41.2% | — |
| Aider Polyglot | — | 60.4% |
| CursorBench | 39.6% | — |
| LMArena WebDev | 1581 | — |
| FrontierSWE | 19.6% | — |
| GSO | — | 1.3% |
| WeirdML | — | 43.7% |
| LiveBench Coding | — | 82.7% |
| CadEval | — | 54% |
| ALE-Bench | 1,270 | — |

## Agentic & Tool Use

- Gemini 3.8 Flash: 40.3 (#22)
- o3-mini: 29.3 (#79)

| Benchmark | Gemini 3.8 Flash | o3-mini |
|---|---|---|
| APEX-Agents | 64.3% | — |
| Cybench | — | 22.5% |
| GDP.pdf | 23.4% | — |
| Vending-Bench 2 | 5,094 | — |

## Reasoning

- Gemini 3.8 Flash: 64.7 (#13)
- o3-mini: 16.3 (#295)

| Benchmark | Gemini 3.8 Flash | o3-mini |
|---|---|---|
| CritPt | 18.3% | 0.3% |
| Chess Puzzles | 61% | 17% |
| LMArena Hard Prompts | 1506 | 1366 |
| Mystery Game Puzzles | 47% | 7% |
| DTBench | 95.7% | 68.8% |
| LMCA | 52.9% | 19% |
| Epoch Capabilities Index | 157.13 | 140.35 |
| ARC-AGI-2 | — | 3% |
| SimpleBench | — | 22.8% |
| NYT Connections (extended) | 97.4% | — |
| ARC-AGI-1 | — | 34.5% |
| LiveBench Reasoning | — | 89.6% |
| LiveBench Data Analysis | — | 70.6% |
| Surface Evolver Bench | 76.9% | — |
| ForecastBench | — | 59.6 |
| LiveBench | — | 75.9% |

## Math

- Gemini 3.8 Flash: 65.2 (#22)
- o3-mini: 27.6 (#236)

| Benchmark | Gemini 3.8 Flash | o3-mini |
|---|---|---|
| FrontierMath (Tiers 1-3) | 68.4% | 18.6% |
| FrontierMath Tier 4 | 22% | 0% |
| OTIS Mock AIME 2024-2025 | 98.9% | 76.9% |
| LMArena Math | 1528 | 1396 |
| ProofBench | 48% | — |
| LiveBench Math | — | 77.3% |
| MATH Level 5 | — | 96.5% |
| FrontierMath (Feb 2025 set) | — | 12.4% |
| FrontierMath Tier 4 (v1) | — | 4.2% |

## Knowledge

- Gemini 3.8 Flash: 74.9 (#2)
- o3-mini: 38.3 (#135)

| Benchmark | Gemini 3.8 Flash | o3-mini |
|---|---|---|
| GPQA Diamond | 95.4% | 77% |
| SimpleQA Verified | 69.7% | 15.3% |
| LMArena Expert | 1529 | 1364 |
| Humanity's Last Exam | 44.5% | — |
| Confabulations | — | 17.9% |

## Multimodal

- Gemini 3.8 Flash: 40.9 (#33)
- o3-mini: —

| Benchmark | Gemini 3.8 Flash | o3-mini |
|---|---|---|
| LMArena Vision | 1312 | — |
| Blueprint-Bench 2 | 38.6% | — |
| Furniture Assembly | 31.7% | — |

## Multilingual

- Gemini 3.8 Flash: 57.9 (#5)
- o3-mini: 45.9 (#159)

| Benchmark | Gemini 3.8 Flash | o3-mini |
|---|---|---|
| LMArena Non-English | 1486 | 1319 |
| LMArena Chinese | 1547 | 1379 |
| LMArena French | 1502 | 1333 |
| LMArena German | 1470 | 1303 |
| LMArena Japanese | 1505 | 1286 |
| LMArena Korean | 1479 | 1314 |
| LMArena Russian | 1507 | 1304 |
| LMArena Spanish | 1492 | 1321 |

## Instruction Following

- Gemini 3.8 Flash: 78.2 (#11)
- o3-mini: 75.3 (#67)

| Benchmark | Gemini 3.8 Flash | o3-mini |
|---|---|---|
| LMArena Instruction Following | 1489 | 1338 |
| LiveBench Instruction Following | — | 84.4% |

## Long Context

- Gemini 3.8 Flash: 46.2 (#23)
- o3-mini: 33.8 (#251)

| Benchmark | Gemini 3.8 Flash | o3-mini |
|---|---|---|
| LMArena Longer Query | 1506 | 1343 |
| Fiction.LiveBench | — | 50% |

## Writing & Preference

- Gemini 3.8 Flash: 72.1 (#16)
- o3-mini: 50.4 (#177)

| Benchmark | Gemini 3.8 Flash | o3-mini |
|---|---|---|
| LMArena Text | 1494 | 1337 |
| LMArena Creative Writing | 1485 | 1286 |
| LMArena Multi-Turn | 1503 | 1320 |
| Short-Story Creative Writing | — | 61.7% |
| EQ-Bench Creative Writing | 1748 | — |
| LiveBench Language | — | 50.7% |

## FAQ

### Is Gemini 3.8 Flash better than o3-mini?

Gemini 3.8 Flash is the stronger model overall, scoring 59.5 to 36.7 on the Noometry Index.

### Which is cheaper, Gemini 3.8 Flash or o3-mini?

Gemini 3.8 Flash is cheaper. It lists at $0.75 per million input tokens and $3.75 per million output tokens; o3-mini lists at $1.10 and $4.40.

### Is Gemini 3.8 Flash or o3-mini better for coding?

Gemini 3.8 Flash scores higher on coding benchmarks: 56.4 versus 41.4 in the Noometry coding category.

### Which has the bigger context window?

Gemini 3.8 Flash does, with 1.05M tokens against 200K.

### How many benchmarks do Gemini 3.8 Flash and o3-mini share?

29 benchmarks have published results for both models. Gemini 3.8 Flash has 46 scored results on Noometry and o3-mini has 51.