GPT-5.6 Luna vs GPT OSS 20B High: Specs & Benchmark Comparison

GPT-5.6 Luna and GPT OSS 20B High both come from OpenAI, GPT OSS 20B High was released in October 2025, and GPT-5.6 Luna followed 9 months later in July 2026. GPT OSS 20B High has a published size of about 20 billion parameters; OpenAI has not disclosed the parameter count of GPT-5.6 Luna.

We have 45 benchmark results for GPT-5.6 Luna and 2 benchmark results for GPT OSS 20B High, but they overlap on a single test: GPQA, where GPT-5.6 Luna scores 92.3% against 74.2%.

GPT OSS 20B High is the cheaper API at $0.10 per million input tokens and $0.50 per million output tokens, roughly 2 times cheaper than GPT-5.6 Luna at $0.20 and $1.20. GPT-5.6 Luna takes the larger context window at 1M tokens, compared with 131K for GPT OSS 20B High. GPT-5.6 Luna accepts text and images as input, while GPT OSS 20B High accepts text.

CharacteristicGPT-5.6 LunaGPT OSS 20B High
CompanyOpenAIOpenAI
Release DateJuly 9, 2026October 1, 2025
Parameters20B
MultimodalYesNo
Context (input)1.1M131K
Context (output)128K131K
Input Price / 1M$0.20$0.10
Output Price / 1M$1.20$0.50
Average Score51.5%86.5%
Benchmarks
GPQA92.3%74.2%

Visual Benchmark Comparison

GPT-5.6 Luna
GPT OSS 20B High
GPQA0.9 vs 0.7
0.9
0.7

Verdict

GPT-5.6 Luna leads in 2 out of 4 comparison categories.

Overall Performance

Both models show comparable average scores: GPT-5.6 Luna — 0.5, GPT OSS 20B High — 0.9.

API Cost

GPT OSS 20B High is 2.3x cheaper: input $0.10/1M vs $0.20/1M tokens.

Context Window

GPT-5.6 Luna supports a larger context: 1M vs 131K tokens.

Recency

GPT-5.6 Luna is newer: released 7/9/2026 vs 10/1/2025.

More About These Models

Frequently Asked Questions

Which is better for coding — GPT-5.6 Luna or GPT OSS 20B High?
Direct comparison on the SWE-Bench benchmark is not available. We recommend reviewing other metrics on the comparison page.
Which model is cheaper — GPT-5.6 Luna or GPT OSS 20B High?
GPT OSS 20B High is cheaper for input: $0.10 per 1M tokens vs $0.20.
Which has a larger context window — GPT-5.6 Luna or GPT OSS 20B High?
GPT-5.6 Luna supports a larger context: 1,050,000 tokens vs 131,072.

The GPT-5.6 Luna and GPT OSS 20B High comparison is updated for 2026. Data includes benchmark results, API pricing, context window size and other specifications. For more detailed information, visit the GPT-5.6 Luna or GPT OSS 20B High page. See also the complete list of AI model comparisons.