token.app › Compare › Gemma 2 27B vs Phi 4

Gemma 2 27B vs Phi 4

Phi 4 leads on 2 of 2 shared benchmarks. Phi 4 is 7.4× cheaper per token on a blended 3:1 basis. Phi 4 accepts the larger context window (16K).

Head to head
Blended price uses a 3:1 input:output mix
Gemma 2 27BPhi 4
ProviderGoogleMicrosoft
Input $/1M$0.650$0.070
Output $/1M$0.650$0.140
Blended $/1M$0.650$0.088
Context window8K16K
Max output2K16K
ReleasedJul 2024Jan 2025
LicenceOpen weightsOpen weights
GPQA Diamond 36.5% 56.1%
OTIS Mock AIME 1.4% 13.8%
Benchmark scores independently run and published by Epoch AI, 'AI Benchmarking Hub', used under CC BY 4.0. Only benchmarks with a score for BOTH models are compared.
Choose Gemma 2 27B if…
  • No clear advantage on the data we hold.
Choose Phi 4 if…
  • Costs less per token — $0.088 vs $0.650 blended
  • Higher GPQA Diamond (56.1% vs 36.5%)
  • Higher OTIS Mock AIME (13.8% vs 1.4%)
  • Larger context window (16K)
  • Newer model (released Jan 2025)
FAQ
Which is better, Gemma 2 27B or Phi 4?
Across the 2 benchmarks both models have independently-run scores for, Gemma 2 27B leads on 0 and Phi 4 leads on 2. Benchmarks are one input — price and context window are on this page too.
Is Gemma 2 27B cheaper than Phi 4?
Phi 4 is cheaper: $0.088 versus $0.650 per million tokens blended at 3:1 input:output. Input alone: $0.650 vs $0.070. Output alone: $0.650 vs $0.140.
What context windows do Gemma 2 27B and Phi 4 have?
Gemma 2 27B: 8,192 tokens. Phi 4: 16,384 tokens.
Who makes Gemma 2 27B and Phi 4?
Gemma 2 27B is made by Google. Phi 4 is made by Microsoft.
Related comparisons