Compare/Gemma 3 27B Instruct vs Granite 3.3 8B (Non-reasoning)

Gemma 3 27B InstructvsGranite 3.3 8B (Non-reasoning)

Side-by-side comparison of pricing, 12 benchmarks, and generation speed.

Google

Gemma 3 27B Instruct

Input
$0.11/M
Output
$0.25/M
Speed
TTFT
IBM

Granite 3.3 8B (Non-reasoning)

Input
$0.03/M
Output
$0.25/M
Speed
380 tok/s
TTFT
20.30s

Winner by Category

Cheaper
Granite 3.3 8B (Non-reasoning)
Faster (tok/s)
Granite 3.3 8B (Non-reasoning)
Lower Latency
Gemma 3 27B Instruct
Benchmarks (12-0)
Gemma 3 27B Instruct

Pricing Comparison

MetricGemma 3 27B InstructGranite 3.3 8B (Non-reasoning)
Input ($/M tokens)$0.11$0.03
Output ($/M tokens)$0.25$0.25
Cost for 1M input + 100K output tokens:
Gemma 3 27B Instruct$0.14
Granite 3.3 8B (Non-reasoning)$0.06

Speed Comparison

Output Speed (tokens/s) — higher is better
Gemma 3 27B Instruct
Granite 3.3 8B (Non-reasoning)
380 tok/s
Time to First Token (seconds) — lower is better
Gemma 3 27B Instruct
Granite 3.3 8B (Non-reasoning)
20.30s

Benchmark Comparison

Data from Artificial Analysis API — 12 benchmarks

Intelligence Index
10.37.0
Coding Index
9.63.4
Math Index
20.76.7
GPQA Diamond
42.8%33.8%
MMLU-Pro
66.9%46.8%
LiveCodeBench
13.7%12.7%
AIME 2025
20.7%6.7%
MATH-500
88.3%66.5%
Humanity's Last Exam
4.7%4.2%
SciCode
21.2%10.1%
IFBench
31.8%22.4%
TerminalBench
3.8%0.0%
Gemma 3 27B Instruct12 wins
0 winsGranite 3.3 8B (Non-reasoning)

Frequently Asked Questions

Which is cheaper, Gemma 3 27B Instruct or Granite 3.3 8B (Non-reasoning)?

Granite 3.3 8B (Non-reasoning) is cheaper overall. Its blended price (3:1 input/output ratio) is $0.09/M tokens vs $0.14/M for Gemma 3 27B Instruct.

Which model performs better on benchmarks?

Gemma 3 27B Instruct wins 12 out of 12 benchmarks compared to 0 for Granite 3.3 8B (Non-reasoning). See the detailed benchmark chart above for per-category results.

Which is faster for real-time applications?

Granite 3.3 8B (Non-reasoning) generates tokens faster at 380 tok/s vs 0 tok/s. Gemma 3 27B Instruct also has lower time-to-first-token (0.00s vs 20.30s).

When should I use Gemma 3 27B Instruct vs Granite 3.3 8B (Non-reasoning)?

Choose based on your priorities: Granite 3.3 8B (Non-reasoning) for lower cost, Gemma 3 27B Instruct for stronger benchmark performance, and Granite 3.3 8B (Non-reasoning) for faster generation. For latency-sensitive apps, check the TTFT comparison above.