# Google Gemma 3 12B

As of October 21, 2026, Google Gemma 3 12B ranks #213 of 213 models overall with 3.8 on the Artificial Analysis Intelligence Index. Its strongest category is math, where it ranks #100 of 116. Google Gemma 3 12B costs $0.05 per million input tokens and $0.15 per million output tokens. It was released on March 13, 2025.

Page: https://llmbenchmark.ai/models/gemma-3-12b-it
Data from October 21, 2026. Every figure below names the board it was read from and the day.

- Maker: Google
- Released: March 13, 2025
- Price per million input tokens: $0.05
- Price per million output tokens: $0.15
- Context window: 131K tokens
- Overall rank: 213 of 213, by the Artificial Analysis Intelligence Index
- Artificial Analysis Intelligence Index: 3.8

Benchmark results (41 of 190):

- AA Intelligence Index: 3.8 (source: Artificial Analysis (via OpenRouter), read October 21, 2026)
- AA Coding Index: 5.8 (source: Artificial Analysis (via OpenRouter), read October 21, 2026)
- AA Agentic Index: 0.1 (source: Artificial Analysis (via OpenRouter), read October 21, 2026)
- AA Math Index: 18.3 (source: Artificial Analysis, read October 21, 2026)
- GPQA Diamond (Artificial Analysis run): 34.9% (source: Artificial Analysis, read October 21, 2026)
- Humanity's Last Exam (Artificial Analysis run): 4.2% (source: Artificial Analysis, read October 21, 2026)
- MMLU Pro (Artificial Analysis run): 59.5% (source: Artificial Analysis, read October 21, 2026)
- LiveCodeBench (Artificial Analysis run): 13.7% (source: Artificial Analysis, read October 21, 2026)
- SciCode: 16.4% (source: Artificial Analysis, read October 21, 2026)
- Terminal-Bench Hard (Artificial Analysis run): 0.8% (source: Artificial Analysis, read October 21, 2026)
- IFBench: 36.7% (source: Artificial Analysis, read October 21, 2026)
- Long context reasoning (AA-LCR): 8.3% (source: Artificial Analysis, read October 21, 2026)
- τ²-Bench Telecom: 10.8% (source: Artificial Analysis, read October 21, 2026)
- AIME 2025 (Artificial Analysis run): 18.3% (source: Artificial Analysis, read October 21, 2026)
- GPQA Diamond (Epoch AI run): 39.5% (source: Epoch AI benchmarking hub, read October 21, 2026)
- OTIS Mock AIME: 16.7% (source: Epoch AI benchmarking hub, read October 21, 2026)
- DTBench: 48.8% (source: Epoch AI benchmarking hub, read October 21, 2026)
- LMCA: 4.5% (source: Epoch AI benchmarking hub, read October 21, 2026)
- Epoch Capabilities Index: 123.5 (source: Epoch AI benchmarking hub, read October 21, 2026)
- LMArena Text: 1334 Elo (source: LMArena (Hugging Face dataset), read October 21, 2026)
- LMArena Text: coding: 1281 Elo (source: LMArena (Hugging Face dataset), read October 21, 2026)
- LMArena Text: hard prompts: 1309 Elo (source: LMArena (Hugging Face dataset), read October 21, 2026)
- LMArena Text: math: 1307 Elo (source: LMArena (Hugging Face dataset), read October 21, 2026)
- LMArena Text: instruction following: 1299 Elo (source: LMArena (Hugging Face dataset), read October 21, 2026)
- LMArena Text: creative writing: 1331 Elo (source: LMArena (Hugging Face dataset), read October 21, 2026)
- LMArena Text: English: 1349 Elo (source: LMArena (Hugging Face dataset), read October 21, 2026)
- LMArena Text: non-English: 1318 Elo (source: LMArena (Hugging Face dataset), read October 21, 2026)
- LMArena Text: German: 1370 Elo (source: LMArena (Hugging Face dataset), read October 21, 2026)
- LMArena Text: Russian: 1335 Elo (source: LMArena (Hugging Face dataset), read October 21, 2026)
- LMArena Text: multi-turn: 1334 Elo (source: LMArena (Hugging Face dataset), read October 21, 2026)
- LMArena Text: longer queries: 1317 Elo (source: LMArena (Hugging Face dataset), read October 21, 2026)
- LMArena Text: expert: 1248 Elo (source: LMArena (Hugging Face dataset), read October 21, 2026)
- LMArena Text: hard prompts in English: 1318 Elo (source: LMArena (Hugging Face dataset), read October 21, 2026)
- LMArena Text: software and IT: 1313 Elo (source: LMArena (Hugging Face dataset), read October 21, 2026)
- LMArena Text: mathematical work: 1336 Elo (source: LMArena (Hugging Face dataset), read October 21, 2026)
- LMArena Text: science: 1366 Elo (source: LMArena (Hugging Face dataset), read October 21, 2026)
- LMArena Text: medicine and healthcare: 1334 Elo (source: LMArena (Hugging Face dataset), read October 21, 2026)
- LMArena Text: legal and government: 1381 Elo (source: LMArena (Hugging Face dataset), read October 21, 2026)
- LMArena Text: business and finance: 1376 Elo (source: LMArena (Hugging Face dataset), read October 21, 2026)
- LMArena Text: writing and language: 1307 Elo (source: LMArena (Hugging Face dataset), read October 21, 2026)
- LMArena Text: entertainment and media: 1297 Elo (source: LMArena (Hugging Face dataset), read October 21, 2026)