# Gemini 3.8 Flash vs Qwen2.5 7B Instruct

> Gemini 3.8 Flash is the stronger model overall, scoring 59.5 to 28.6 on the Noometry Index. Qwen2.5 7B Instruct costs 4.9× less per token, which makes it the better buy when Gemini 3.8 Flash's lead doesn't matter for your workload.

- Canonical page: https://noometry.com/compare/gemini-3-8-flash-vs-qwen2-5-7b-instruct
- Last updated: 2026-10-10
- Shared benchmarks: 6

## Summary

- They share 6 benchmarks with published results for both. Gemini 3.8 Flash scores higher in 7 categories and Qwen2.5 7B Instruct in 0 categories; 7 gaps are clear of the uncertainty.
- The widest gap is in knowledge, where Gemini 3.8 Flash leads 74.9 to 17.0.
- The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 98.9% for Gemini 3.8 Flash and 2.5% for Qwen2.5 7B Instruct.
- Qwen2.5 7B Instruct is cheaper at $0.17 / $0.70 per million input/output tokens, against $0.75 / $3.75 for Gemini 3.8 Flash.
- Gemini 3.8 Flash accepts more context: 1.05M tokens versus 131K.
- Qwen2.5 7B Instruct has downloadable open weights; the other is API-only.

## Snapshot

| | Gemini 3.8 Flash | Qwen2.5 7B Instruct |
|---|---|---|
| Provider | Google | Alibaba (Qwen) |
| Noometry Index | 59.5 | 28.6 |
| Rank | 13 | 314 |
| Context | 1.05M | 131K |
| Input $/M | $0.75 | $0.17 |
| Output $/M | $3.75 | $0.70 |
| Weights | Proprietary | Open |

## Coding

- Gemini 3.8 Flash: 56.4 (#17)
- Qwen2.5 7B Instruct: 35.9 (#204)

| Benchmark | Gemini 3.8 Flash | Qwen2.5 7B Instruct |
|---|---|---|
| DeepSWE | 73.8% | — |
| FrontierCode | 41.2% | — |
| CursorBench | 39.6% | — |
| LMArena WebDev | 1581 | — |
| FrontierSWE | 19.6% | — |
| SciCode | 56.6% | — |
| BigCodeBench Instruct | — | 37.6% |
| LMArena Coding | 1512 | — |
| BigCodeBench Complete | — | 46.1% |
| ALE-Bench | 1,270 | — |

## Agentic & Tool Use

- Gemini 3.8 Flash: 40.3 (#22)
- Qwen2.5 7B Instruct: 23.5 (#117)

| Benchmark | Gemini 3.8 Flash | Qwen2.5 7B Instruct |
|---|---|---|
| APEX-Agents | 64.3% | — |
| BALROG | — | 7.8% |
| GDP.pdf | 23.4% | — |
| Vending-Bench 2 | 5,094 | — |

## Reasoning

- Gemini 3.8 Flash: 64.7 (#13)
- Qwen2.5 7B Instruct: 14.3 (#317)

| Benchmark | Gemini 3.8 Flash | Qwen2.5 7B Instruct |
|---|---|---|
| Chess Puzzles | 61% | 0% |
| DTBench | 95.7% | 47.7% |
| LMCA | 52.9% | 6.4% |
| Epoch Capabilities Index | 157.13 | 118.44 |
| NYT Connections (extended) | 97.4% | — |
| CritPt | 18.3% | — |
| LMArena Hard Prompts | 1506 | — |
| Mystery Game Puzzles | 47% | — |
| Surface Evolver Bench | 76.9% | — |

## Math

- Gemini 3.8 Flash: 65.2 (#22)
- Qwen2.5 7B Instruct: 11.8 (#300)

| Benchmark | Gemini 3.8 Flash | Qwen2.5 7B Instruct |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 98.9% | 2.5% |
| FrontierMath (Tiers 1-3) | 68.4% | — |
| FrontierMath Tier 4 | 22% | — |
| ProofBench | 48% | — |
| Omni-MATH | — | 29.4% |
| LMArena Math | 1528 | — |

## Knowledge

- Gemini 3.8 Flash: 74.9 (#2)
- Qwen2.5 7B Instruct: 17.0 (#281)

| Benchmark | Gemini 3.8 Flash | Qwen2.5 7B Instruct |
|---|---|---|
| GPQA Diamond | 95.4% | 35.5% |
| Humanity's Last Exam | 44.5% | — |
| SimpleQA Verified | 69.7% | — |
| MMLU-Pro | — | 53.9% |
| GPQA (HELM) | — | 34.1% |
| LMArena Expert | 1529 | — |
| MMLU | — | 72.9% |

## Multimodal

- Gemini 3.8 Flash: 40.9 (#33)
- Qwen2.5 7B Instruct: —

| Benchmark | Gemini 3.8 Flash | Qwen2.5 7B Instruct |
|---|---|---|
| LMArena Vision | 1312 | — |
| Blueprint-Bench 2 | 38.6% | — |
| Furniture Assembly | 31.7% | — |

## Multilingual

- Gemini 3.8 Flash: 57.9 (#5)
- Qwen2.5 7B Instruct: —

| Benchmark | Gemini 3.8 Flash | Qwen2.5 7B Instruct |
|---|---|---|
| LMArena Non-English | 1486 | — |
| LMArena Chinese | 1547 | — |
| LMArena French | 1502 | — |
| LMArena German | 1470 | — |
| LMArena Japanese | 1505 | — |
| LMArena Korean | 1479 | — |
| LMArena Russian | 1507 | — |
| LMArena Spanish | 1492 | — |

## Instruction Following

- Gemini 3.8 Flash: 78.2 (#11)
- Qwen2.5 7B Instruct: 63.4 (#226)

| Benchmark | Gemini 3.8 Flash | Qwen2.5 7B Instruct |
|---|---|---|
| IFEval | — | 74.1% |
| LMArena Instruction Following | 1489 | — |

## Long Context

- Gemini 3.8 Flash: 46.2 (#23)
- Qwen2.5 7B Instruct: —

| Benchmark | Gemini 3.8 Flash | Qwen2.5 7B Instruct |
|---|---|---|
| LMArena Longer Query | 1506 | — |

## Writing & Preference

- Gemini 3.8 Flash: 72.1 (#16)
- Qwen2.5 7B Instruct: 48.8 (#194)

| Benchmark | Gemini 3.8 Flash | Qwen2.5 7B Instruct |
|---|---|---|
| LMArena Text | 1494 | — |
| LMArena Creative Writing | 1485 | — |
| EQ-Bench Creative Writing | 1748 | — |
| WildBench | — | 73.1% |
| LMArena Multi-Turn | 1503 | — |

## FAQ

### Is Gemini 3.8 Flash better than Qwen2.5 7B Instruct?

Gemini 3.8 Flash is the stronger model overall, scoring 59.5 to 28.6 on the Noometry Index. Qwen2.5 7B Instruct costs 4.9× less per token, which makes it the better buy when Gemini 3.8 Flash's lead doesn't matter for your workload.

### Which is cheaper, Gemini 3.8 Flash or Qwen2.5 7B Instruct?

Qwen2.5 7B Instruct is cheaper. It lists at $0.17 per million input tokens and $0.70 per million output tokens; Gemini 3.8 Flash lists at $0.75 and $3.75.

### Is Gemini 3.8 Flash or Qwen2.5 7B Instruct better for coding?

Gemini 3.8 Flash scores higher on coding benchmarks: 56.4 versus 35.9 in the Noometry coding category.

### Which has the bigger context window?

Gemini 3.8 Flash does, with 1.05M tokens against 131K.

### How many benchmarks do Gemini 3.8 Flash and Qwen2.5 7B Instruct share?

6 benchmarks have published results for both models. Gemini 3.8 Flash has 46 scored results on Noometry and Qwen2.5 7B Instruct has 15.