token.app › Compare › Phi 4 vs gpt-oss-120b

Phi 4 vs gpt-oss-120b

gpt-oss-120b leads on 2 of 2 shared benchmarks. gpt-oss-120b is 1.2× cheaper per token on a blended 3:1 basis. gpt-oss-120b accepts the larger context window (131K).

Head to head
Blended price uses a 3:1 input:output mix
Phi 4gpt-oss-120b
ProviderMicrosoftOpenAI
Input $/1M$0.070$0.037
Output $/1M$0.140$0.170
Blended $/1M$0.088$0.070
Context window16K131K
Max output16K131K
ReleasedJan 2025Aug 2025
LicenceOpen weightsOpen weights
GPQA Diamond 56.1% 75.8%
OTIS Mock AIME 13.8% 88.9%
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 Phi 4 if…
  • No clear advantage on the data we hold.
Choose gpt-oss-120b if…
  • Costs less per token — $0.070 vs $0.088 blended
  • Higher GPQA Diamond (75.8% vs 56.1%)
  • Higher OTIS Mock AIME (88.9% vs 13.8%)
  • Larger context window (131K)
  • Newer model (released Aug 2025)
FAQ
Which is better, Phi 4 or gpt-oss-120b?
Across the 2 benchmarks both models have independently-run scores for, Phi 4 leads on 0 and gpt-oss-120b leads on 2. Benchmarks are one input — price and context window are on this page too.
Is Phi 4 cheaper than gpt-oss-120b?
gpt-oss-120b is cheaper: $0.070 versus $0.088 per million tokens blended at 3:1 input:output. Input alone: $0.070 vs $0.037. Output alone: $0.140 vs $0.170.
What context windows do Phi 4 and gpt-oss-120b have?
Phi 4: 16,384 tokens. gpt-oss-120b: 131,072 tokens.
Who makes Phi 4 and gpt-oss-120b?
Phi 4 is made by Microsoft. gpt-oss-120b is made by OpenAI.
Related comparisons