GPT OSS 20B vs GPT 5.6 Luna
Compare GPT OSS 20B and GPT 5.6 Luna on key metrics including price, context length, throughput, and other model features.
AuthorOpenAI
Context Length131.1k
Supports Tools
OpenAI’s 21B-parameter open-weight Mixture-of-Experts (MoE) model, released under the Apache 2.0 license. Features 3.6B active parameters per forward pass, optimized for low-latency inference and deployability on consumer or single-GPU hardware. Trained in OpenAI’s Harmony response format, it supports reasoning level configuration, fine-tuning, and agentic capabilities such as function calling and structured outputs.
Activity
Last 14 daysPrompt
2BCompletion
69MTotal
2BStartup
Startup
Latency (p50)0.23s
Throughput (p50)176.6 tok/s
Pricing
Input$0.009/M tokens
Output$0.04/M tokens
Cached input$0.0045/M tokens
Features
Input Modalitiestext, file
Output Modalitiestext
Supported EndpointsChat Completions
Vision
Supports Tools
AuthorOpenAI
Context Length1.1M
Supports Tools
GPT-5.6 Luna is a fast, cost-efficient model in OpenAI's GPT-5.6 series. It is suited for high-volume, latency-sensitive tasks such as chat, classification, and lightweight agentic workflows, providing capable reasoning for its price tier.
Activity
Last 14 daysPrompt
199MCompletion
5MTotal
204MStartup
Startup
Latency (p50)1.06s
Throughput (p50)73.7 tok/s
Pricing
Input$0.05/M tokens
Output$0.30/M tokens
Cached input$0.005/M tokens
Features
Input Modalitiestext, image, file
Output Modalitiestext
Supported EndpointsChat Completions
Vision
Supports Tools