gpt-oss-20b
OPENAI Developer Architecture Profile
Intelligence (ELO)1039Chatbot Arena Verified
Max Context Limit131,072Max Out: 32,768 tokens
Prompt Caching50% OFF$0.0090 / 1M cached
Standard API / 1M$0.11Blended Standard Rate
Model Capabilities & Contract SLAs
gpt-oss-20b is an open-weight 21B parameter model released by OpenAI under the Apache 2.0 license. It uses a Mixture-of-Experts (MoE) architecture with 3.6B active parameters per forward pass, optimized for...
Granular Pricing Matrix
Input Tokens (Standard Prompt)$0.02 / 1M
⚡ Cached Input (Prompt Cache Read)50% OFF$0.0090 / 1M
Output Tokens (Completion)$0.09 / 1M
Pricing data via OpenRouter. Sync: 10/2/2026
Evaluate Competitors
VS Engine Matchupgpt-oss-20b vs Inference.net: Schematron V2 TurboVS Engine Matchupgpt-oss-20b vs OpenRouter: FusionVS Engine Matchupgpt-oss-20b vs Xiaomi: MiMo-V2.5VS Engine Matchupgpt-oss-20b vs OpenAI: gpt-oss-20b (batch)VS Engine Matchupgpt-oss-20b vs Microsoft: Phi 4VS Engine Matchupgpt-oss-20b vs TypeSafe: Jev Router