Model Comparison

GPT-5.6 Luna vs GPT-5.6 SolWhich is better in 2026?

GPT-5.6 Sol significantly outperforms across most benchmarks. GPT-5.6 Luna is 5.0x cheaper per token.

Verdict: GPT-5.6 Luna vs GPT-5.6 Sol — which is better?

GPT-5.6 Luna (by OpenAI) and GPT-5.6 Sol (by OpenAI) are two of the AI models people compare most. Here is how they stack up on benchmarks, price and capabilities, and which one to pick in 2026.

GPT-5.6 Luna outperforms in 0 benchmarks, while GPT-5.6 Sol is better at 44 benchmarks (Agents' Last Exam, ARC-AGI-3, Artificial Analysis, Artificial Analysis Coding Agent Index v1.1, AutomationBench, BenchCAD, BenchCAD (with Python tool), Big Finance Bench, BrowseComp, Capture-the-Flag Challenges (Internal), Connectors, DeepSWE, ExploitBench, ExploitGym, FrontierMath, FrontierMath Tier 4 (v2), GDP.pdf, GDPval-AA, GeneBench-Pro, GPQA, Graphwalks BFS >128k, Graphwalks BFS 1M, HealthBench, HealthBench Consensus, HealthBench Hard, HealthBench Professional, Internal Research Debugging Evaluation, KernelGen 1P, LifeSciBench, Management Consulting Tasks (Internal), MedChemBench (Internal), MMMU-Pro, MMMU-Pro (with tools), MRCR v2 (8-needle), MRCR v2 (8-needle, 512K-1M), NanoGPT, OSWorld 2.0, PostTrainBench Lite, RSI Index, Search and Function-Calling, SEC-bench Pro, SWE-Bench Pro, Terminal-Bench 2.1, Toolathlon). GPT-5.6 Sol significantly outperforms across most benchmarks.

On price, GPT-5.6 Luna is roughly 5.0x cheaper per token on a blended 3:1 input/output basis, which adds up quickly at production volume.

Choose GPT-5.6 Luna if…

  • cost matters — it's about 5.0x cheaper per token

Choose GPT-5.6 Sol if…

  • you want the strongest raw capability — it leads on 44 of 44 shared benchmarks

Performance Benchmarks

Comparative analysis across standard metrics

44 benchmarks

GPT-5.6 Luna outperforms in 0 benchmarks, while GPT-5.6 Sol is better at 44 benchmarks (Agents' Last Exam, ARC-AGI-3, Artificial Analysis, Artificial Analysis Coding Agent Index v1.1, AutomationBench, BenchCAD, BenchCAD (with Python tool), Big Finance Bench, BrowseComp, Capture-the-Flag Challenges (Internal), Connectors, DeepSWE, ExploitBench, ExploitGym, FrontierMath, FrontierMath Tier 4 (v2), GDP.pdf, GDPval-AA, GeneBench-Pro, GPQA, Graphwalks BFS >128k, Graphwalks BFS 1M, HealthBench, HealthBench Consensus, HealthBench Hard, HealthBench Professional, Internal Research Debugging Evaluation, KernelGen 1P, LifeSciBench, Management Consulting Tasks (Internal), MedChemBench (Internal), MMMU-Pro, MMMU-Pro (with tools), MRCR v2 (8-needle), MRCR v2 (8-needle, 512K-1M), NanoGPT, OSWorld 2.0, PostTrainBench Lite, RSI Index, Search and Function-Calling, SEC-bench Pro, SWE-Bench Pro, Terminal-Bench 2.1, Toolathlon).

GPT-5.6 Sol significantly outperforms across most benchmarks.

Fri Jul 17 2026 • llm-stats.com

Arena Performance

Human preference votes

Pricing Analysis

Price comparison per million tokens

GPT-5.6 Luna costs less

For input processing, GPT-5.6 Luna ($1.00/1M tokens) is 5.0x cheaper than GPT-5.6 Sol ($5.00/1M tokens).

For output processing, GPT-5.6 Luna ($6.00/1M tokens) is 5.0x cheaper than GPT-5.6 Sol ($30.00/1M tokens).

In conclusion, GPT-5.6 Sol is more expensive than GPT-5.6 Luna.*

* Using a 3:1 ratio of input to output tokens

Lowest available price from all providers
Fri Jul 17 2026 • llm-stats.com
OpenAI
GPT-5.6 Luna
Input tokens$1.00
Output tokens$6.00
Best providerOpenAI
OpenAI
GPT-5.6 Sol
Input tokens$5.00
Output tokens$30.00
Best providerOpenAI
Notice missing or incorrect data?Start an Issue→

Context Window

Maximum input and output token capacity

Both models have the same input context window of 1,050,000 tokens. Both models can generate responses up to 128,000 tokens.

OpenAI
GPT-5.6 Luna
Input1,050,000 tokens
Output128,000 tokens
OpenAI
GPT-5.6 Sol
Input1,050,000 tokens
Output128,000 tokens
Fri Jul 17 2026 • llm-stats.com

Input Capabilities

Supported data types and modalities

Both GPT-5.6 Luna and GPT-5.6 Sol support multimodal inputs.

They are both capable of processing various types of data, offering versatility in application.

GPT-5.6 Luna

Text
Images
Audio
Video

GPT-5.6 Sol

Text
Images
Audio
Video

License

Usage and distribution terms

Both models are licensed under proprietary licenses.

Both models have usage restrictions defined by their respective organizations.

GPT-5.6 Luna

Proprietary

Closed source

GPT-5.6 Sol

Proprietary

Closed source

Release Timeline

When each model was launched

Both models were released on 2026-07-09.

They likely represent similar generations of model development.

GPT-5.6 Luna

Jul 9, 2026

1 weeks ago

GPT-5.6 Sol

Jul 9, 2026

1 weeks ago

Knowledge Cutoff

When training data ends

Both models have the same knowledge cutoff date of 2026-02-16.

They should have similar awareness of historical events and information up to this date.

GPT-5.6 Luna

Feb 2026

GPT-5.6 Sol

Feb 2026

Provider Availability

GPT-5.6 Luna is available from OpenAI. GPT-5.6 Sol is available from OpenAI.

GPT-5.6 Luna

openai logo
OpenAI
Input Price:Input: $1.00/1MOutput Price:Output: $6.00/1M

GPT-5.6 Sol

openai logo
OpenAI
Input Price:Input: $5.00/1MOutput Price:Output: $30.00/1M
* Prices shown are per million tokens

Outputs Comparison

Notice missing or incorrect data?Start an Issue discussion→

Key Takeaways

OpenAI
Less expensive input tokens
Less expensive output tokens
OpenAI
Higher Agents' Last Exam score (52.7% vs 50.3%)
Higher ARC-AGI-3 score (7.8% vs 0.2%)
Higher Artificial Analysis score (59.0% vs 51.0%)
Higher Artificial Analysis Coding Agent Index v1.1 score (80.0% vs 74.6%)
Higher AutomationBench score (18.1% vs 14.9%)
Higher BenchCAD score (70.6% vs 63.1%)
Higher BenchCAD (with Python tool) score (83.4% vs 73.9%)
Higher Big Finance Bench score (53.0% vs 36.0%)
Higher BrowseComp score (90.4% vs 83.3%)
Higher Capture-the-Flag Challenges (Internal) score (96.7% vs 85.2%)
Higher Connectors score (100.0% vs 99.9%)
Higher DeepSWE score (72.7% vs 67.2%)
Higher ExploitBench score (73.5% vs 33.2%)
Higher ExploitGym score (33.7% vs 12.4%)
Higher FrontierMath score (89.0% vs 78.6%)
Higher FrontierMath Tier 4 (v2) score (83.0% vs 58.5%)
Higher GDP.pdf score (30.7% vs 22.7%)
Higher GDPval-AA score (58.3% vs 53.1%)
Higher GeneBench-Pro score (28.7% vs 10.8%)
Higher GPQA score (94.6% vs 92.3%)
Higher Graphwalks BFS >128k score (90.7% vs 81.3%)
Higher Graphwalks BFS 1M score (77.1% vs 51.2%)
Higher HealthBench score (57.0% vs 55.8%)
Higher HealthBench Consensus score (95.5% vs 95.1%)
Higher HealthBench Hard score (33.1% vs 32.0%)
Higher HealthBench Professional score (60.5% vs 55.7%)
Higher Internal Research Debugging Evaluation score (68.3% vs 50.8%)
Higher KernelGen 1P score (61.1% vs 22.4%)
Higher LifeSciBench score (59.9% vs 51.2%)
Higher Management Consulting Tasks (Internal) score (43.2% vs 35.4%)
Higher MedChemBench (Internal) score (48.3% vs 30.4%)
Higher MMMU-Pro score (83.0% vs 78.4%)
Higher MMMU-Pro (with tools) score (84.6% vs 79.5%)
Higher MRCR v2 (8-needle) score (91.5% vs 41.3%)
Higher MRCR v2 (8-needle, 512K-1M) score (73.8% vs 41.3%)
Higher NanoGPT score (9.7% vs 1.7%)
Higher OSWorld 2.0 score (62.6% vs 45.6%)
Higher PostTrainBench Lite score (50.3% vs 29.6%)
Higher RSI Index score (57.9% vs 41.9%)
Higher Search and Function-Calling score (91.0% vs 89.7%)
Higher SEC-bench Pro score (71.2% vs 48.9%)
Higher SWE-Bench Pro score (64.6% vs 62.7%)
Higher Terminal-Bench 2.1 score (88.8% vs 84.7%)
Higher Toolathlon score (58.0% vs 53.4%)

Detailed Comparison

Interactive Arena

Judge for yourself.

Run your own prompts against GPT-5.6 Luna and GPT-5.6 Sol side-by-side, then vote on the output you prefer.

GPT-5.6 Luna
✓ Preferred
GPT-5.6 Sol
Open in Playground
AI Model Comparison Table
Feature
OpenAI
GPT-5.6 Luna
OpenAI
GPT-5.6 Sol

FAQ

Common questions about GPT-5.6 Luna vs GPT-5.6 Sol.

Which is better, GPT-5.6 Luna or GPT-5.6 Sol?

GPT-5.6 Sol significantly outperforms across most benchmarks. GPT-5.6 Luna is made by OpenAI and GPT-5.6 Sol is made by OpenAI. The best choice depends on your use case — compare their benchmark scores, pricing, and capabilities above.

How does GPT-5.6 Luna compare to GPT-5.6 Sol in benchmarks?

GPT-5.6 Luna scores Connectors: 99.9%, HealthBench Consensus: 95.1%, GPQA: 92.3%, Search and Function-Calling: 89.7%, Capture-the-Flag Challenges (Internal): 85.2%. GPT-5.6 Sol scores Connectors: 100.0%, Capture-the-Flag Challenges (Internal): 96.7%, HealthBench Consensus: 95.5%, GPQA: 94.6%, MRCR v2 (8-needle): 91.5%.

Is GPT-5.6 Luna cheaper than GPT-5.6 Sol?

GPT-5.6 Luna is 5.0x cheaper for input tokens. GPT-5.6 Luna costs $1.00/M input and $6.00/M output via openai. GPT-5.6 Sol costs $5.00/M input and $30.00/M output via openai.

What are the context window sizes for GPT-5.6 Luna and GPT-5.6 Sol?

GPT-5.6 Luna supports 1.1M tokens and GPT-5.6 Sol supports 1.1M tokens. A larger context window lets you process longer documents, conversations, or codebases in a single request.

What are the main differences between GPT-5.6 Luna and GPT-5.6 Sol?

Key differences include input pricing ($1.00 vs $5.00/M). See the full comparison above for benchmark-by-benchmark results.

Related comparisons