# Gemini 2.5 Flash-Lite vs Hy4 preview

> Hy4 preview is the stronger model overall, scoring 44.6 to 36.4 on the Noometry Index. Gemini 2.5 Flash-Lite costs 7.1× less per token, which makes it the better buy when Hy4 preview's lead doesn't matter for your workload.

- Canonical page: https://noometry.com/compare/gemini-2-5-flash-lite-vs-hy4-preview
- Last updated: 2026-10-10
- Shared benchmarks: 0

## Summary

- The widest gap is in math, where Hy4 preview leads 54.5 to 35.5.
- Gemini 2.5 Flash-Lite is cheaper at $0.10 / $0.40 per million input/output tokens, against $0.83 / $2.50 for Hy4 preview.
- Hy4 preview has downloadable open weights; the other is API-only.

## Snapshot

| | Gemini 2.5 Flash-Lite | Hy4 preview |
|---|---|---|
| Provider | Google | Tencent |
| Noometry Index | 36.4 | 44.6 |
| Rank | 208 | 73 |
| Context | 1.05M | 1.05M |
| Input $/M | $0.10 | $0.83 |
| Output $/M | $0.40 | $2.50 |
| Weights | Proprietary | Open |

## Coding

- Gemini 2.5 Flash-Lite: 38.0 (#167)
- Hy4 preview: 51.4 (#36)

| Benchmark | Gemini 2.5 Flash-Lite | Hy4 preview |
|---|---|---|
| LMArena WebDev | — | 1633 |
| WeirdML | 35.2% | — |
| LMArena Coding | 1373 | — |
| ALE-Bench | 325.9 | — |

## Agentic & Tool Use

- Gemini 2.5 Flash-Lite: 27.8 (#89)
- Hy4 preview: —

| Benchmark | Gemini 2.5 Flash-Lite | Hy4 preview |
|---|---|---|
| Berkeley Function Calling Leaderboard | 36.9% | — |

## Reasoning

- Gemini 2.5 Flash-Lite: 21.5 (#196)
- Hy4 preview: 30.6 (#75)

| Benchmark | Gemini 2.5 Flash-Lite | Hy4 preview |
|---|---|---|
| Kagi LLM Benchmark | 40.5% | — |
| NYT Connections (extended) | — | 68.2% |
| LMArena Hard Prompts | 1377 | — |
| DTBench | 62.8% | — |
| LMCA | 18.1% | — |
| Epoch Capabilities Index | 133.93 | — |

## Math

- Gemini 2.5 Flash-Lite: 35.5 (#144)
- Hy4 preview: 54.5 (#37)

| Benchmark | Gemini 2.5 Flash-Lite | Hy4 preview |
|---|---|---|
| ProofBench | — | 75% |
| Omni-MATH | 48% | — |
| LMArena Math | 1373 | — |

## Knowledge

- Gemini 2.5 Flash-Lite: 32.1 (#206)
- Hy4 preview: —

| Benchmark | Gemini 2.5 Flash-Lite | Hy4 preview |
|---|---|---|
| MMLU-Pro | 53.7% | — |
| Vectara Hallucination Rate | 3.3% | — |
| GPQA (HELM) | 30.9% | — |
| LMArena Expert | 1373 | — |

## Multimodal

- Gemini 2.5 Flash-Lite: 28.6 (#109)
- Hy4 preview: —

| Benchmark | Gemini 2.5 Flash-Lite | Hy4 preview |
|---|---|---|
| LMArena Vision | 1198 | — |
| VPCT | 30% | — |

## Multilingual

- Gemini 2.5 Flash-Lite: 49.5 (#129)
- Hy4 preview: —

| Benchmark | Gemini 2.5 Flash-Lite | Hy4 preview |
|---|---|---|
| LMArena Non-English | 1369 | — |
| LMArena Chinese | 1404 | — |
| LMArena French | 1388 | — |
| LMArena German | 1390 | — |
| LMArena Japanese | 1359 | — |
| LMArena Korean | 1358 | — |
| LMArena Russian | 1373 | — |
| LMArena Spanish | 1396 | — |

## Instruction Following

- Gemini 2.5 Flash-Lite: 70.2 (#163)
- Hy4 preview: —

| Benchmark | Gemini 2.5 Flash-Lite | Hy4 preview |
|---|---|---|
| IFEval | 81% | — |
| LMArena Instruction Following | 1366 | — |

## Long Context

- Gemini 2.5 Flash-Lite: 33.3 (#257)
- Hy4 preview: —

| Benchmark | Gemini 2.5 Flash-Lite | Hy4 preview |
|---|---|---|
| Fiction.LiveBench | 47.2% | — |
| LMArena Longer Query | 1373 | — |

## Writing & Preference

- Gemini 2.5 Flash-Lite: 57.0 (#130)
- Hy4 preview: —

| Benchmark | Gemini 2.5 Flash-Lite | Hy4 preview |
|---|---|---|
| LMArena Text | 1379 | — |
| LMArena Creative Writing | 1366 | — |
| WildBench | 81.8% | — |
| LMArena Multi-Turn | 1367 | — |

## FAQ

### Is Gemini 2.5 Flash-Lite better than Hy4 preview?

Hy4 preview is the stronger model overall, scoring 44.6 to 36.4 on the Noometry Index. Gemini 2.5 Flash-Lite costs 7.1× less per token, which makes it the better buy when Hy4 preview's lead doesn't matter for your workload.

### Which is cheaper, Gemini 2.5 Flash-Lite or Hy4 preview?

Gemini 2.5 Flash-Lite is cheaper. It lists at $0.10 per million input tokens and $0.40 per million output tokens; Hy4 preview lists at $0.83 and $2.50.

### Is Gemini 2.5 Flash-Lite or Hy4 preview better for coding?

Hy4 preview scores higher on coding benchmarks: 51.4 versus 38.0 in the Noometry coding category.

### Which has the bigger context window?

Both accept 1.05M tokens.

### How many benchmarks do Gemini 2.5 Flash-Lite and Hy4 preview share?

0 benchmarks have published results for both models. Gemini 2.5 Flash-Lite has 33 scored results on Noometry and Hy4 preview has 3.