Gemma 2 27B
GoogleLLMs
83.5
Performance
★ 4.3
Rating
245
Reviews
ReasoningMediumCode AssistantText GenerationOpen Weight
About
Google's open-weight 27B model offering strong performance for developers who want a capable model with permissive commercial terms.
Strengths
Strong on knowledge and reasoning for its size. Permissive commercial license from Google. Good for developers wanting a capable open model. Competitive with much larger models on MMLU and HumanEval.
Specifications
- Context window
- 8,192
- Parameters
- 27B
Pricing
- Input cost
- Free
- Output cost
- Free
Open-weight with permissive Gemma license. Free to download.
Speed & Latency
- 95
- tokens/sec
- 200ms
- time to first token
Available On
HuggingFaceGoogle AI StudioOllamaKaggle
Features
streamingsystem messages
Market Sentiment
Polymarket prediction market odds for "Best AI Model" — showing how the market values each provider over time.
Loading sentiment data...
Performance Trend
Benchmark score trends over time for the top 5 benchmarks.
Loading history...
Benchmarks
Scores from various benchmark tests; higher is better.
| Test | Score | Percentile | Source |
|---|---|---|---|
| BigBench Hard | 76.0 | p86 | seed |
| DROP | 77.0 | p86 | seed |
| GSM8K | 88.0 | p93 | seed |
| HumanEval | 82.0 | p91 |
