# Gemini 2.5 Flash vs Mixtral 8x22B

> Gemini 2.5 Flash is the stronger model overall, scoring 39.3 to 27.1 on the Noometry Index.

- Canonical page: https://noometry.com/compare/gemini-2-5-flash-vs-mixtral-8x22b
- Last updated: 2026-10-10
- Shared benchmarks: 26

## Summary

- They share 26 benchmarks with published results for both. Gemini 2.5 Flash scores higher in 8 categories and Mixtral 8x22B in 1 category; 9 gaps are clear of the uncertainty.
- The widest gap is in knowledge, where Gemini 2.5 Flash leads 36.4 to 15.1.
- The biggest single-benchmark swing is WeirdML: 41.9% for Gemini 2.5 Flash and 3.2% for Mixtral 8x22B.
- Gemini 2.5 Flash is cheaper at $0.30 / $2.50 per million input/output tokens, against $2 / $6 for Mixtral 8x22B.
- Gemini 2.5 Flash accepts more context: 1.05M tokens versus 64K.
- Mixtral 8x22B has downloadable open weights; the other is API-only.

## Snapshot

| | Gemini 2.5 Flash | Mixtral 8x22B |
|---|---|---|
| Provider | Google | Mistral AI |
| Noometry Index | 39.3 | 27.1 |
| Rank | 170 | 333 |
| Context | 1.05M | 64K |
| Input $/M | $0.30 | $2 |
| Output $/M | $2.50 | $6 |
| Weights | Proprietary | Open |

## Coding

- Gemini 2.5 Flash: 35.8 (#220)
- Mixtral 8x22B: 24.2 (#329)

| Benchmark | Gemini 2.5 Flash | Mixtral 8x22B |
|---|---|---|
| WeirdML | 41.9% | 3.2% |
| LMArena Coding | 1424 | 1166 |
| SWE-bench Verified (bash only) | 28.7% | — |
| Aider Polyglot | 55.1% | — |
| BigCodeBench Instruct | — | 40.6% |
| BigCodeBench Complete | — | 50.2% |
| ALE-Bench | 661.88 | — |
| HumanEval+ | — | 72% |
| MBPP+ | — | 64.3% |

## Agentic & Tool Use

- Gemini 2.5 Flash: 30.8 (#74)
- Mixtral 8x22B: 23.1 (#127)

| Benchmark | Gemini 2.5 Flash | Mixtral 8x22B |
|---|---|---|
| Terminal-Bench | 17.1% | — |
| Berkeley Function Calling Leaderboard | 56.2% | — |
| TheAgentCompany | 41.1% | — |
| Cybench | — | 7.5% |
| BALROG | 33.5% | — |
| Vending-Bench 2 | 548.84 | — |

## Reasoning

- Gemini 2.5 Flash: 18.1 (#286)
- Mixtral 8x22B: 19.9 (#248)

| Benchmark | Gemini 2.5 Flash | Mixtral 8x22B |
|---|---|---|
| LMArena Hard Prompts | 1422 | 1150 |
| DTBench | 76.5% | 55.1% |
| Epoch Capabilities Index | 143.03 | 122.03 |
| ForecastBench | 60.6 | 56.3 |
| ARC-AGI-2 | 2.5% | — |
| SimpleBench | 41.2% | — |
| Kagi LLM Benchmark | 56.8% | — |
| ARC-AGI-1 | 33.3% | — |
| CritPt | 1.1% | — |
| EnigmaEval | 2.7% | — |
| LMCA | 27.5% | — |

## Math

- Gemini 2.5 Flash: 39.9 (#98)
- Mixtral 8x22B: 22.9 (#275)

| Benchmark | Gemini 2.5 Flash | Mixtral 8x22B |
|---|---|---|
| Omni-MATH | 38.5% | 16.3% |
| LMArena Math | 1415 | 1184 |
| OTIS Mock AIME 2024-2025 | 73.1% | — |
| MATH Level 5 | — | 24.2% |
| FrontierMath (Feb 2025 set) | 4.8% | — |
| FrontierMath Tier 4 (v1) | 4.2% | — |

## Knowledge

- Gemini 2.5 Flash: 36.4 (#168)
- Mixtral 8x22B: 15.1 (#293)

| Benchmark | Gemini 2.5 Flash | Mixtral 8x22B |
|---|---|---|
| MMLU-Pro | 63.9% | 46% |
| GPQA (HELM) | 39% | 33.4% |
| LMArena Expert | 1426 | 1113 |
| GPQA Diamond | — | 34.1% |
| Humanity's Last Exam | 12.1% | — |
| Confabulations | 16.8% | — |
| Vectara Hallucination Rate | 7.8% | — |
| MMLU | — | 77.8% |

## Multimodal

- Gemini 2.5 Flash: 41.8 (#32)
- Mixtral 8x22B: —

| Benchmark | Gemini 2.5 Flash | Mixtral 8x22B |
|---|---|---|
| LMArena Vision | 1253 | — |
| GeoBench | 76% | — |
| VPCT | 46.2% | — |
| SpatialViz-Bench | 36.9% | — |

## Multilingual

- Gemini 2.5 Flash: 52.3 (#88)
- Mixtral 8x22B: 32.8 (#255)

| Benchmark | Gemini 2.5 Flash | Mixtral 8x22B |
|---|---|---|
| LMArena Non-English | 1409 | 1128 |
| LMArena Chinese | 1450 | 1116 |
| LMArena French | 1433 | 1166 |
| LMArena German | 1418 | 1141 |
| LMArena Japanese | 1405 | 1037 |
| LMArena Korean | 1385 | 1057 |
| LMArena Russian | 1415 | 1158 |
| LMArena Spanish | 1421 | 1151 |

## Instruction Following

- Gemini 2.5 Flash: 75.7 (#54)
- Mixtral 8x22B: 57.7 (#266)

| Benchmark | Gemini 2.5 Flash | Mixtral 8x22B |
|---|---|---|
| IFEval | 89.8% | 72.4% |
| LMArena Instruction Following | 1405 | 1147 |

## Long Context

- Gemini 2.5 Flash: 47.5 (#17)
- Mixtral 8x22B: 34.7 (#247)

| Benchmark | Gemini 2.5 Flash | Mixtral 8x22B |
|---|---|---|
| LMArena Longer Query | 1419 | 1144 |
| Fiction.LiveBench | 77.8% | — |

## Writing & Preference

- Gemini 2.5 Flash: 53.8 (#157)
- Mixtral 8x22B: 36.9 (#262)

| Benchmark | Gemini 2.5 Flash | Mixtral 8x22B |
|---|---|---|
| LMArena Text | 1417 | 1162 |
| LMArena Creative Writing | 1400 | 1141 |
| WildBench | 81.7% | 71.1% |
| LMArena Multi-Turn | 1408 | 1130 |
| Short-Story Creative Writing | 76.5% | — |
| EQ-Bench Creative Writing | 1137 | — |

## FAQ

### Is Gemini 2.5 Flash better than Mixtral 8x22B?

Gemini 2.5 Flash is the stronger model overall, scoring 39.3 to 27.1 on the Noometry Index.

### Which is cheaper, Gemini 2.5 Flash or Mixtral 8x22B?

Gemini 2.5 Flash is cheaper. It lists at $0.30 per million input tokens and $2.50 per million output tokens; Mixtral 8x22B lists at $2 and $6.

### Is Gemini 2.5 Flash or Mixtral 8x22B better for coding?

Gemini 2.5 Flash scores higher on coding benchmarks: 35.8 versus 24.2 in the Noometry coding category.

### Which has the bigger context window?

Gemini 2.5 Flash does, with 1.05M tokens against 64K.

### How many benchmarks do Gemini 2.5 Flash and Mixtral 8x22B share?

26 benchmarks have published results for both models. Gemini 2.5 Flash has 54 scored results on Noometry and Mixtral 8x22B has 34.