gpt-oss-120b vs gpt-5-nano-2025-08-07
Pricing, Performance & Features Comparison
OpenAI most powerful open weight model, which fits into a single H100 GPU.
Input–
Output–
Latency (p50)–
Output Limit131K
Function Calling
JSON Mode
-
InputText
OutputText
–
Context Length400K
Reasoning
Providers1
ReleasedAug 2025
Knowledge Cutoff–
License–
Superseded Bygpt-5.4-nano
GPT-5 Nano is our fastest, cheapest version of GPT-5. It's great for summarization and classification tasks.
Input$0.5
Output$0.4
Latency (p50)5.4s
Output Limit128K
Function Calling
JSON Mode
-
InputText, Image
OutputText
in$0.5out$0.4–write$0.05