Side by side

See the difference.

Scores, capabilities and costs for the models you’re considering.

Comparison of gpt-oss-20b
1 modelopenaigpt-oss-20bRemove ×
What’s differentgpt-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…
DateAug 4, 2025Listed
Listed rate$0.018 in / $0.09 out/ 1M tokensDarkbloom · via OpenRouter · all routes ↗darkbloom/fp8 · quantization=fp8
Provider endpoints20 endpoints →
Context window131.1K
Max output32.8K
Parameters20B
Tool callingSupported
Structured outputSupported
ReasoningSupported
Prompt cachingSupported
Open weightsSupported

Add another model

Search all models →