# Gemini 2.5 Flash-Lite vs Qwen3 Max

> Qwen3 Max is the stronger model overall, scoring 43.5 to 36.4 on the Noometry Index. Gemini 2.5 Flash-Lite costs 14× less per token, which makes it the better buy when Qwen3 Max's lead doesn't matter for your workload.

- Canonical page: https://noometry.com/compare/gemini-2-5-flash-lite-vs-qwen3-max
- Last updated: 2026-10-10
- Shared benchmarks: 23

## Summary

- They share 23 benchmarks with published results for both. Gemini 2.5 Flash-Lite scores higher in 0 categories and Qwen3 Max in 8 categories; 7 gaps are clear of the uncertainty.
- The widest gap is in knowledge, where Qwen3 Max leads 48.1 to 32.1.
- The biggest single-benchmark swing is Kagi LLM Benchmark: 40.5% for Gemini 2.5 Flash-Lite and 72.5% for Qwen3 Max.
- Gemini 2.5 Flash-Lite is cheaper at $0.10 / $0.40 per million input/output tokens, against $1.20 / $6 for Qwen3 Max.
- Gemini 2.5 Flash-Lite accepts more context: 1.05M tokens versus 262K.

## Snapshot

| | Gemini 2.5 Flash-Lite | Qwen3 Max |
|---|---|---|
| Provider | Google | Alibaba (Qwen) |
| Noometry Index | 36.4 | 43.5 |
| Rank | 208 | 82 |
| Context | 1.05M | 262K |
| Input $/M | $0.10 | $1.20 |
| Output $/M | $0.40 | $6 |
| Weights | Proprietary | Proprietary |

## Coding

- Gemini 2.5 Flash-Lite: 38.0 (#167)
- Qwen3 Max: 42.4 (#93)

| Benchmark | Gemini 2.5 Flash-Lite | Qwen3 Max |
|---|---|---|
| LMArena Coding | 1373 | 1456 |
| ALE-Bench | 325.9 | 370.45 |
| WeirdML | 35.2% | — |

## Agentic & Tool Use

- Gemini 2.5 Flash-Lite: 27.8 (#89)
- Qwen3 Max: —

| Benchmark | Gemini 2.5 Flash-Lite | Qwen3 Max |
|---|---|---|
| Berkeley Function Calling Leaderboard | 36.9% | — |
| Vending-Bench 2 | — | 71.56 |

## Reasoning

- Gemini 2.5 Flash-Lite: 21.5 (#196)
- Qwen3 Max: 22.0 (#186)

| Benchmark | Gemini 2.5 Flash-Lite | Qwen3 Max |
|---|---|---|
| Kagi LLM Benchmark | 40.5% | 72.5% |
| LMArena Hard Prompts | 1377 | 1448 |
| DTBench | 62.8% | 82.1% |
| LMCA | 18.1% | 28.3% |
| Epoch Capabilities Index | 133.93 | 142.41 |
| NYT Connections (extended) | — | 30.1% |
| Chess Puzzles | — | 4% |
| Mystery Game Puzzles | — | 5% |

## Math

- Gemini 2.5 Flash-Lite: 35.5 (#144)
- Qwen3 Max: 37.6 (#94)

| Benchmark | Gemini 2.5 Flash-Lite | Qwen3 Max |
|---|---|---|
| LMArena Math | 1373 | 1447 |
| FrontierMath (Tiers 1-3) | — | 18.9% |
| OTIS Mock AIME 2024-2025 | — | 73.3% |
| Omni-MATH | 48% | — |
| MATH Level 5 | — | 97.1% |

## Knowledge

- Gemini 2.5 Flash-Lite: 32.1 (#206)
- Qwen3 Max: 48.1 (#72)

| Benchmark | Gemini 2.5 Flash-Lite | Qwen3 Max |
|---|---|---|
| LMArena Expert | 1373 | 1457 |
| GPQA Diamond | — | 72.6% |
| SimpleQA Verified | — | 48.7% |
| MMLU-Pro | 53.7% | — |
| Vectara Hallucination Rate | 3.3% | — |
| GPQA (HELM) | 30.9% | — |

## Multimodal

- Gemini 2.5 Flash-Lite: 28.6 (#109)
- Qwen3 Max: —

| Benchmark | Gemini 2.5 Flash-Lite | Qwen3 Max |
|---|---|---|
| LMArena Vision | 1198 | — |
| VPCT | 30% | — |

## Multilingual

- Gemini 2.5 Flash-Lite: 49.5 (#129)
- Qwen3 Max: 53.9 (#58)

| Benchmark | Gemini 2.5 Flash-Lite | Qwen3 Max |
|---|---|---|
| LMArena Non-English | 1369 | 1429 |
| LMArena Chinese | 1404 | 1479 |
| LMArena French | 1388 | 1450 |
| LMArena German | 1390 | 1463 |
| LMArena Japanese | 1359 | 1399 |
| LMArena Korean | 1358 | 1397 |
| LMArena Russian | 1373 | 1428 |
| LMArena Spanish | 1396 | 1463 |

## Instruction Following

- Gemini 2.5 Flash-Lite: 70.2 (#163)
- Qwen3 Max: 75.0 (#81)

| Benchmark | Gemini 2.5 Flash-Lite | Qwen3 Max |
|---|---|---|
| LMArena Instruction Following | 1366 | 1420 |
| IFEval | 81% | — |

## Long Context

- Gemini 2.5 Flash-Lite: 33.3 (#257)
- Qwen3 Max: 41.6 (#129)

| Benchmark | Gemini 2.5 Flash-Lite | Qwen3 Max |
|---|---|---|
| Fiction.LiveBench | 47.2% | 66.7% |
| LMArena Longer Query | 1373 | 1439 |
| CL-bench | — | 14.5% |

## Writing & Preference

- Gemini 2.5 Flash-Lite: 57.0 (#130)
- Qwen3 Max: 62.6 (#72)

| Benchmark | Gemini 2.5 Flash-Lite | Qwen3 Max |
|---|---|---|
| LMArena Text | 1379 | 1439 |
| LMArena Creative Writing | 1366 | 1401 |
| LMArena Multi-Turn | 1367 | 1446 |
| WildBench | 81.8% | — |

## FAQ

### Is Gemini 2.5 Flash-Lite better than Qwen3 Max?

Qwen3 Max is the stronger model overall, scoring 43.5 to 36.4 on the Noometry Index. Gemini 2.5 Flash-Lite costs 14× less per token, which makes it the better buy when Qwen3 Max's lead doesn't matter for your workload.

### Which is cheaper, Gemini 2.5 Flash-Lite or Qwen3 Max?

Gemini 2.5 Flash-Lite is cheaper. It lists at $0.10 per million input tokens and $0.40 per million output tokens; Qwen3 Max lists at $1.20 and $6.

### Is Gemini 2.5 Flash-Lite or Qwen3 Max better for coding?

Qwen3 Max scores higher on coding benchmarks: 42.4 versus 38.0 in the Noometry coding category.

### Which has the bigger context window?

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

### How many benchmarks do Gemini 2.5 Flash-Lite and Qwen3 Max share?

23 benchmarks have published results for both models. Gemini 2.5 Flash-Lite has 33 scored results on Noometry and Qwen3 Max has 33.