Price Per TokenPrice Per Token
Google
Google
vs
Google
Google

Gemma 2 27B vs Gemma 3 12B

A detailed comparison of pricing, benchmarks, and capabilities

OpenClaw

Best LLMs for OpenClaw Vote for which model works best with OpenClaw

112 out of our 301 tracked models have had a price change in February.

Get our weekly newsletter on pricing changes, new releases, and tools.

Key Takeaways

Gemma 2 27B wins:

  • Higher intelligence benchmark

Gemma 3 12B wins:

  • Cheaper input tokens
  • Cheaper output tokens
  • Larger context window
  • Faster response time
  • Better at coding
  • Better at math
  • Supports vision
  • Supports tool calls
Price Advantage
Gemma 3 12B
Benchmark Advantage
Gemma 3 12B
Context Window
Gemma 3 12B
Speed
Gemma 3 12B

Pricing Comparison

Price Comparison

MetricGemma 2 27BGemma 3 12BWinner
Input (per 1M tokens)$0.65$0.03 Gemma 3 12B
Output (per 1M tokens)$0.65$0.10 Gemma 3 12B
Cache Read (per 1M)N/A$15000.00 Gemma 3 12B
Using a 3:1 input/output ratio, Gemma 3 12B is 93% cheaper overall.

Gemma 2 27B Providers

NextBit $0.65 (Cheapest)

Gemma 3 12B Providers

Chutes $0.03 (Cheapest)
DeepInfra $0.04
Crusoe $0.08
Cloudflare $0.35

Benchmark Comparison

9
Benchmarks Compared
1
Gemma 2 27B Wins
2
Gemma 3 12B Wins

Benchmark Scores

BenchmarkGemma 2 27BGemma 3 12BWinner
Intelligence Index
Overall intelligence score
36.28.8
Coding Index
Code generation & understanding
-6.3-
Math Index
Mathematical reasoning
-18.3-
MMLU Pro
Academic knowledge
38.459.5
GPQA
Graduate-level science
16.734.9
LiveCodeBench
Competitive programming
-13.7-
Aider
Real-world code editing
36.1--
AIME
Competition math
-22.0-
BBH
Big-Bench Hard
49.3--
Gemma 3 12B significantly outperforms in coding benchmarks.

Cost vs Quality

X-axis:
Y-axis:
Loading chart...
Gemma 2 27B
Other models

Context & Performance

Context Window

Gemma 2 27B
8,192
tokens
Max output: 2,048 tokens
Gemma 3 12B
131,072
tokens
Max output: 131,072 tokens
Gemma 3 12B has a 94% larger context window.

Speed Performance

MetricGemma 2 27BGemma 3 12BWinner
Tokens/secondN/A33.6 tok/s
Time to First TokenN/A11.94s

Capabilities

Feature Comparison

FeatureGemma 2 27BGemma 3 12B
Vision (Image Input)
Tool/Function Calls
Reasoning Mode
Audio Input
Audio Output
PDF Input
Prompt Caching
Web Search

License & Release

PropertyGemma 2 27BGemma 3 12B
LicenseOpen SourceOpen Source
AuthorGoogleGoogle
ReleasedJul 2024Mar 2025

Gemma 2 27B Modalities

Input
text
Output
text

Gemma 3 12B Modalities

Input
textimage
Output
text

Related Comparisons

Compare Gemma 2 27B with:

Compare Gemma 3 12B with:

Frequently Asked Questions

Gemma 3 12B has cheaper input pricing at $0.03/M tokens. Gemma 3 12B has cheaper output pricing at $0.10/M tokens.
Gemma 3 12B scores higher on coding benchmarks with a score of 6.3, compared to Gemma 2 27B's score of N/A.
Gemma 2 27B has a 8,192 token context window, while Gemma 3 12B has a 131,072 token context window.
Gemma 2 27B does not support vision. Gemma 3 12B supports vision.