# Gemma 3 12B vs Qwen3 8B

> Qwen3 8B is the stronger model overall, scoring 33.3 to 31.8 on the Noometry Index. Gemma 3 12B costs 4.1× less per token, which makes it the better buy when Qwen3 8B's lead doesn't matter for your workload.

- Canonical page: https://noometry.com/compare/gemma-3-12b-vs-qwen3-8b
- Last updated: 2026-10-10
- Shared benchmarks: 10

## Summary

- They share 10 benchmarks with published results for both. Gemma 3 12B scores higher in 1 category and Qwen3 8B in 5 categories; 5 gaps are clear of the uncertainty.
- The widest gap is in math, where Qwen3 8B leads 33.3 to 21.0.
- The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 16.7% for Gemma 3 12B and 56.1% for Qwen3 8B.
- Gemma 3 12B is cheaper at $0.05 / $0.15 per million input/output tokens, against $0.18 / $0.70 for Qwen3 8B.

## Snapshot

| | Gemma 3 12B | Qwen3 8B |
|---|---|---|
| Provider | Google | Alibaba (Qwen) |
| Noometry Index | 31.8 | 33.3 |
| Rank | 255 | 237 |
| Context | 131K | 131K |
| Input $/M | $0.05 | $0.18 |
| Output $/M | $0.15 | $0.70 |
| Weights | Open | Open |

## Coding

- Gemma 3 12B: 31.2 (#276)
- Qwen3 8B: 33.4 (#245)

| Benchmark | Gemma 3 12B | Qwen3 8B |
|---|---|---|
| SciCode | 17.4% | 22.6% |
| LMArena Coding | 1281 | — |

## Agentic & Tool Use

- Gemma 3 12B: 25.3 (#101)
- Qwen3 8B: 29.9 (#71)

| Benchmark | Gemma 3 12B | Qwen3 8B |
|---|---|---|
| Berkeley Function Calling Leaderboard | 30.4% | 42.6% |

## Reasoning

- Gemma 3 12B: 15.2 (#307)
- Qwen3 8B: 16.1 (#299)

| Benchmark | Gemma 3 12B | Qwen3 8B |
|---|---|---|
| CritPt | 0% | 0% |
| Chess Puzzles | 0% | 5% |
| DTBench | 48.8% | 59.7% |
| LMCA | 4.5% | 8.8% |
| Epoch Capabilities Index | 123.49 | 136.17 |
| LMArena Hard Prompts | 1309 | — |

## Math

- Gemma 3 12B: 21.0 (#273)
- Qwen3 8B: 33.3 (#176)

| Benchmark | Gemma 3 12B | Qwen3 8B |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 16.7% | 56.1% |
| LMArena Math | 1307 | — |

## Knowledge

- Gemma 3 12B: 26.5 (#250)
- Qwen3 8B: 36.1 (#164)

| Benchmark | Gemma 3 12B | Qwen3 8B |
|---|---|---|
| GPQA Diamond | 39.5% | 56.8% |
| Vectara Hallucination Rate | 4.4% | 4.8% |
| LMArena Expert | 1248 | — |

## Multimodal

- Gemma 3 12B: —
- Qwen3 8B: —

| Benchmark | Gemma 3 12B | Qwen3 8B |
|---|---|---|
| MindCube | 46.7% | — |

## Multilingual

- Gemma 3 12B: 45.9 (#160)
- Qwen3 8B: —

| Benchmark | Gemma 3 12B | Qwen3 8B |
|---|---|---|
| LMArena Non-English | 1318 | — |
| LMArena German | 1370 | — |
| LMArena Russian | 1335 | — |

## Instruction Following

- Gemma 3 12B: 68.8 (#179)
- Qwen3 8B: —

| Benchmark | Gemma 3 12B | Qwen3 8B |
|---|---|---|
| LMArena Instruction Following | 1299 | — |

## Long Context

- Gemma 3 12B: 40.0 (#157)
- Qwen3 8B: 37.8 (#205)

| Benchmark | Gemma 3 12B | Qwen3 8B |
|---|---|---|
| Fiction.LiveBench | — | 62.1% |
| LMArena Longer Query | 1317 | — |

## Writing & Preference

- Gemma 3 12B: 47.7 (#205)
- Qwen3 8B: —

| Benchmark | Gemma 3 12B | Qwen3 8B |
|---|---|---|
| LMArena Text | 1334 | — |
| LMArena Creative Writing | 1331 | — |
| EQ-Bench Creative Writing | 1126 | — |
| LMArena Multi-Turn | 1334 | — |

## FAQ

### Is Gemma 3 12B better than Qwen3 8B?

Qwen3 8B is the stronger model overall, scoring 33.3 to 31.8 on the Noometry Index. Gemma 3 12B costs 4.1× less per token, which makes it the better buy when Qwen3 8B's lead doesn't matter for your workload.

### Which is cheaper, Gemma 3 12B or Qwen3 8B?

Gemma 3 12B is cheaper. It lists at $0.05 per million input tokens and $0.15 per million output tokens; Qwen3 8B lists at $0.18 and $0.70.

### Is Gemma 3 12B or Qwen3 8B better for coding?

Qwen3 8B scores higher on coding benchmarks: 33.4 versus 31.2 in the Noometry coding category.

### Which has the bigger context window?

Both accept 131K tokens.

### How many benchmarks do Gemma 3 12B and Qwen3 8B share?

10 benchmarks have published results for both models. Gemma 3 12B has 24 scored results on Noometry and Qwen3 8B has 11.