# Gemini 2.5 Flash vs Gemma 4 26B A4B IT

> Gemma 4 26B A4B IT is the stronger model overall, scoring 43.1 to 39.0 on the Noometry Index.

- Canonical page: https://noometry.com/compare/gemini-2-5-flash-vs-gemma-4-26b-a4b-it
- Last updated: 2026-10-10
- Shared benchmarks: 24

## Summary

- They share 24 benchmarks with published results for both. Gemini 2.5 Flash scores higher in 3 categories and Gemma 4 26B A4B IT in 6 categories; 7 gaps are clear of the uncertainty.
- The widest gap is in knowledge, where Gemma 4 26B A4B IT leads 45.6 to 36.1.
- The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 73.1% for Gemini 2.5 Flash and 82.2% for Gemma 4 26B A4B IT.
- Gemma 4 26B A4B IT is cheaper at $0.0765 / $0.26 per million input/output tokens, against $0.30 / $2.50 for Gemini 2.5 Flash.
- Gemini 2.5 Flash accepts more context: 1.05M tokens versus 262K.
- Gemma 4 26B A4B IT has downloadable open weights; the other is API-only.

## Snapshot

| | Gemini 2.5 Flash | Gemma 4 26B A4B IT |
|---|---|---|
| Provider | Google | Google |
| Noometry Index | 39.0 | 43.1 |
| Rank | 162 | 88 |
| Context | 1.05M | 262K |
| Input $/M | $0.30 | $0.0765 |
| Output $/M | $2.50 | $0.26 |
| Weights | Proprietary | Open |

## Coding

- Gemini 2.5 Flash: 35.5 (#210)
- Gemma 4 26B A4B IT: 38.8 (#156)

| Benchmark | Gemini 2.5 Flash | Gemma 4 26B A4B IT |
|---|---|---|
| WeirdML | 41.9% | 35.2% |
| LMArena Coding | 1424 | 1446 |
| ALE-Bench | 661.88 | 927.17 |
| SWE-bench Verified (bash only) | 28.7% | — |
| Aider Polyglot | 55.1% | — |
| LMArena WebDev | — | 1359 |
| SciCode | — | 40% |

## Agentic & Tool Use

- Gemini 2.5 Flash: 30.6 (#67)
- Gemma 4 26B A4B IT: —

| Benchmark | Gemini 2.5 Flash | Gemma 4 26B A4B IT |
|---|---|---|
| Terminal-Bench | 17.1% | — |
| Berkeley Function Calling Leaderboard | 56.2% | — |
| TheAgentCompany | 41.1% | — |
| BALROG | 33.5% | — |
| Vending-Bench 2 | 548.84 | — |

## Reasoning

- Gemini 2.5 Flash: 18.2 (#271)
- Gemma 4 26B A4B IT: 21.1 (#205)

| Benchmark | Gemini 2.5 Flash | Gemma 4 26B A4B IT |
|---|---|---|
| CritPt | 1.1% | 0% |
| LMArena Hard Prompts | 1422 | 1438 |
| DTBench | 76.5% | 74.9% |
| LMCA | 27.5% | 29.7% |
| Epoch Capabilities Index | 143.04 | 141.85 |
| ARC-AGI-2 | 2.5% | — |
| SimpleBench | 41.2% | — |
| Kagi LLM Benchmark | 56.8% | — |
| ARC-AGI-1 | 33.3% | — |
| Chess Puzzles | — | 6% |
| EnigmaEval | 2.7% | — |
| ForecastBench | 60.6 | — |

## Math

- Gemini 2.5 Flash: 38.1 (#87)
- Gemma 4 26B A4B IT: 45.7 (#61)

| Benchmark | Gemini 2.5 Flash | Gemma 4 26B A4B IT |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 73.1% | 82.2% |
| LMArena Math | 1415 | 1469 |
| Omni-MATH | 38.5% | — |
| FrontierMath (Feb 2025 set) | 4.8% | — |
| FrontierMath Tier 4 (v1) | 4.2% | — |

## Knowledge

- Gemini 2.5 Flash: 36.1 (#162)
- Gemma 4 26B A4B IT: 45.6 (#79)

| Benchmark | Gemini 2.5 Flash | Gemma 4 26B A4B IT |
|---|---|---|
| Vectara Hallucination Rate | 7.8% | 5.2% |
| LMArena Expert | 1426 | 1447 |
| GPQA Diamond | — | 73.2% |
| Humanity's Last Exam | 12.1% | — |
| MMLU-Pro | 63.9% | — |
| Confabulations | 16.8% | — |
| GPQA (HELM) | 39% | — |

## Multimodal

- Gemini 2.5 Flash: 41.2 (#31)
- Gemma 4 26B A4B IT: 39.9 (#43)

| Benchmark | Gemini 2.5 Flash | Gemma 4 26B A4B IT |
|---|---|---|
| LMArena Vision | 1254 | 1260 |
| GeoBench | 76% | — |
| VPCT | 46.2% | — |
| SpatialViz-Bench | 36.9% | — |

## Multilingual

- Gemini 2.5 Flash: 52.5 (#83)
- Gemma 4 26B A4B IT: 53.3 (#68)

| Benchmark | Gemini 2.5 Flash | Gemma 4 26B A4B IT |
|---|---|---|
| LMArena Non-English | 1410 | 1422 |
| LMArena Chinese | 1450 | 1495 |
| LMArena French | 1433 | 1459 |
| LMArena Russian | 1415 | 1434 |
| LMArena Spanish | 1421 | 1415 |
| LMArena German | 1419 | — |
| LMArena Japanese | 1403 | — |
| LMArena Korean | 1385 | — |

## Instruction Following

- Gemini 2.5 Flash: 75.9 (#50)
- Gemma 4 26B A4B IT: 75.0 (#78)

| Benchmark | Gemini 2.5 Flash | Gemma 4 26B A4B IT |
|---|---|---|
| LMArena Instruction Following | 1405 | 1420 |
| IFEval | 89.8% | — |

## Long Context

- Gemini 2.5 Flash: 47.5 (#16)
- Gemma 4 26B A4B IT: 43.6 (#87)

| Benchmark | Gemini 2.5 Flash | Gemma 4 26B A4B IT |
|---|---|---|
| LMArena Longer Query | 1419 | 1427 |
| Fiction.LiveBench | 77.8% | — |

## Writing & Preference

- Gemini 2.5 Flash: 54.0 (#153)
- Gemma 4 26B A4B IT: 58.8 (#110)

| Benchmark | Gemini 2.5 Flash | Gemma 4 26B A4B IT |
|---|---|---|
| LMArena Text | 1417 | 1434 |
| LMArena Creative Writing | 1400 | 1401 |
| EQ-Bench Creative Writing | 1137 | 1305 |
| LMArena Multi-Turn | 1408 | 1442 |
| Short-Story Creative Writing | 76.5% | — |
| WildBench | 81.7% | — |

## FAQ

### Is Gemini 2.5 Flash better than Gemma 4 26B A4B IT?

Gemma 4 26B A4B IT is the stronger model overall, scoring 43.1 to 39.0 on the Noometry Index.

### Which is cheaper, Gemini 2.5 Flash or Gemma 4 26B A4B IT?

Gemma 4 26B A4B IT is cheaper. It lists at $0.0765 per million input tokens and $0.26 per million output tokens; Gemini 2.5 Flash lists at $0.30 and $2.50.

### Is Gemini 2.5 Flash or Gemma 4 26B A4B IT better for coding?

Gemma 4 26B A4B IT scores higher on coding benchmarks: 38.8 versus 35.5 in the Noometry coding category.

### Which has the bigger context window?

Gemini 2.5 Flash does, with 1.05M tokens against 262K.

### How many benchmarks do Gemini 2.5 Flash and Gemma 4 26B A4B IT share?

24 benchmarks have published results for both models. Gemini 2.5 Flash has 54 scored results on Noometry and Gemma 4 26B A4B IT has 28.