# Google Gemma 4 26B A4B

As of October 22, 2026, Google Gemma 4 26B A4B ranks #57 of 123 models overall with 54.7 ±17.9 on the LLMBenchmark Index. Gemma 4 26B A4B ranks #30 of 108 in math, its strongest category. Google Gemma 4 26B A4B costs $0.15 per million input tokens and $0.60 per million output tokens. Gemma 4 26B A4B was released on April 3, 2026.

Page: https://llmbenchmark.ai/models/gemma-4-26b-a4b-it
Data from October 22, 2026. Every figure below names the board it was read from and the day.

- Maker: Google
- Released: April 3, 2026
- Price per million input tokens: $0.15
- Price per million output tokens: $0.60
- Context window: 262K tokens
- Overall rank: 57 of 123, by the LLMBenchmark Index
- LLMBenchmark Index: 54.7 ±17.9 (90% margin, from 4 of 7 inputs)
- Artificial Analysis Intelligence Index: 16.7
- Vibe Coding Index: not enough data (one index benchmark; at least two are needed to rank)

Benchmark results (55 of 195):

- LMArena WebDev: 1360 Elo (source: LMArena WebDev, read October 22, 2026)
- AA Intelligence Index: 16.7 (source: Artificial Analysis (via OpenRouter), read October 22, 2026)
- AA Coding Index: 39.3 (source: Artificial Analysis (via OpenRouter), read October 22, 2026)
- AA Agentic Index: 2.9 (source: Artificial Analysis (via OpenRouter), read October 22, 2026)
- GPQA Diamond (Artificial Analysis run): 79.2% (source: Artificial Analysis, read October 22, 2026)
- Humanity's Last Exam (Artificial Analysis run): 19.3% (source: Artificial Analysis, read October 22, 2026)
- SciCode: 40.0% (source: Artificial Analysis, read October 22, 2026)
- Terminal-Bench Hard (Artificial Analysis run): 25.0% (source: Artificial Analysis, read October 22, 2026)
- IFBench: 72.4% (source: Artificial Analysis, read October 22, 2026)
- Long context reasoning (AA-LCR): 65.7% (source: Artificial Analysis, read October 22, 2026)
- τ²-Bench Telecom: 43.6% (source: Artificial Analysis, read October 22, 2026)
- WeirdML: 35.2% (source: Epoch AI benchmarking hub, read October 22, 2026)
- GPQA Diamond (Epoch AI run): 73.2% (source: Epoch AI benchmarking hub, read October 22, 2026)
- OTIS Mock AIME: 82.2% (source: Epoch AI benchmarking hub, read October 22, 2026)
- DTBench: 74.9% (source: Epoch AI benchmarking hub, read October 22, 2026)
- LMCA: 29.7% (source: Epoch AI benchmarking hub, read October 22, 2026)
- ALE-Bench: 927.2 (source: ALE-Bench (Sakana AI), read October 22, 2026)
- Epoch Capabilities Index: 141.9 (source: Epoch AI benchmarking hub, read October 22, 2026)
- LMArena Text: 1434 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026)
- LMArena Text: coding: 1446 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026)
- LMArena Text: hard prompts: 1438 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026)
- LMArena Text: math: 1470 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026)
- LMArena Text: instruction following: 1420 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026)
- LMArena Text: creative writing: 1400 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026)
- LMArena Vision: 1260 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026)
- LMArena Text: English: 1444 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026)
- LMArena Text: non-English: 1421 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026)
- LMArena Text: Chinese: 1495 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026)
- LMArena Text: French: 1460 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026)
- LMArena Text: Spanish: 1415 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026)
- LMArena Text: Russian: 1434 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026)
- LMArena Text: multi-turn: 1441 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026)
- LMArena Text: longer queries: 1427 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026)
- LMArena Text: expert: 1447 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026)
- LMArena Text: hard prompts in English: 1448 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026)
- LMArena Text: software and IT: 1454 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026)
- LMArena Text: mathematical work: 1475 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026)
- LMArena Text: science: 1446 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026)
- LMArena Text: medicine and healthcare: 1428 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026)
- LMArena Text: legal and government: 1457 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026)
- LMArena Text: business and finance: 1429 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026)
- LMArena Text: writing and language: 1405 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026)
- LMArena Text: entertainment and media: 1403 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026)
- LMArena WebDev: React: 1351 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026)
- LMArena WebDev: HTML: 1362 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026)
- LMArena Vision: English: 1256 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026)
- LMArena Vision: Chinese: 1312 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026)
- LMArena Vision: reading text: 1268 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026)
- LMArena Vision: diagrams: 1264 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026)
- LMArena Vision: homework: 1293 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026)
- LMArena Vision: humor: 1272 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026)
- LMArena Vision: captioning: 1297 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026)
- LMArena Vision: recognizing things: 1191 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026)
- LMArena Vision: creative writing: 1256 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026)
- FACTS Benchmark Suite: 42.7% (source: Kaggle Benchmarks, read October 22, 2026)