token.app › Compare › Inkling vs GPT-5.6 Luna
Inkling vs GPT-5.6 Luna
GPT-5.6 Luna leads on 4 of 4 shared benchmarks. GPT-5.6 Luna is 3.9× cheaper per token on a blended 3:1 basis. GPT-5.6 Luna accepts the larger context window (1.1M).
Head to head
Blended price uses a 3:1 input:output mix
| Inkling | GPT-5.6 Luna | |
|---|---|---|
| Provider | Thinkingmachines | OpenAI |
| Input $/1M | $1.00 | $0.200 |
| Output $/1M | $4.05 | $1.20 |
| Blended $/1M | $1.76 | $0.450 |
| Context window | 524K | 1.1M |
| Max output | 472K | 128K |
| Released | Jul 2026 | Jul 2026 |
| Licence | Open weights | Proprietary |
| GPQA Diamond | 88.3% | 91.6% |
| FrontierMath T1–3 | 33.3% | 82.1% |
| FrontierMath T4 | 4.9% | 61.0% |
| OTIS Mock AIME | 88.9% | 98.3% |
Benchmark scores independently run and published by Epoch AI, 'AI Benchmarking Hub', used under CC BY 4.0. Only benchmarks with a score for BOTH models are compared.
Choose Inkling if…
- Open weights — self-hostable
- Newer model (released Jul 2026)
Choose GPT-5.6 Luna if…
- Costs less per token — $0.450 vs $1.76 blended
- Higher GPQA Diamond (91.6% vs 88.3%)
- Higher FrontierMath T1–3 (82.1% vs 33.3%)
- Higher FrontierMath T4 (61.0% vs 4.9%)
- Higher OTIS Mock AIME (98.3% vs 88.9%)
- Larger context window (1.1M)
FAQ
Which is better, Inkling or GPT-5.6 Luna?
Across the 4 benchmarks both models have independently-run scores for, Inkling leads on 0 and GPT-5.6 Luna leads on 4. Benchmarks are one input — price and context window are on this page too.
Is Inkling cheaper than GPT-5.6 Luna?
GPT-5.6 Luna is cheaper: $0.450 versus $1.76 per million tokens blended at 3:1 input:output. Input alone: $1.00 vs $0.200. Output alone: $4.05 vs $1.20.
What context windows do Inkling and GPT-5.6 Luna have?
Inkling: 524,288 tokens. GPT-5.6 Luna: 1,050,000 tokens.
Who makes Inkling and GPT-5.6 Luna?
Inkling is made by Thinkingmachines. GPT-5.6 Luna is made by OpenAI.