text model comparison
GPT-6 Luna vs GPT-6 Sol
GPT-6 Luna and GPT-6 Sol are both current text models, from OpenAI. GPT-6 Luna is 95% cheaper per 1M input tokens than GPT-6 Sol, $0.10 vs $2. On Arena Text, GPT-6 Luna scores 1443 Elo and GPT-6 Sol 1456 Elo (Arena, measured October 8, 2026). Both accept 1,050,000 tokens of context.
Data verified October 12, 2026 · Dataset 2026-10-12
Side by side
Standard-tier prices in effect today. The better value in each row is highlighted.
| Fact | GPT-6 Luna | GPT-6 Sol |
|---|---|---|
| Price per 1M input tokens | $0.10 (better) | $2 |
| Price per 1M output tokens | $0.50 (better) | $10 |
| Cost of 10M input and 2M output tokens | $2 (better) | $40 |
| Arena Text | 1443 Elo (#31 of 39) | 1456 Elo (#26 of 39) (better) |
| GPQA Diamond | 90.5% (#23 of 37) | 94.3% (#7 of 37) (better) |
| FrontierMath Tiers 1-3 (v2) | 78.9% (#14 of 37) | 89.8% (#5 of 37) (better) |
| SimpleQA Verified | 41.4% (#25 of 34) | 60.7% (#14 of 34) (better) |
| Provider | OpenAI | OpenAI |
| Release date | September 22, 2026 | September 22, 2026 |
| Access | API only | API only |
| Context window | 1,050,000 tokens | 1,050,000 tokens |
| Max output | 128,000 tokens | 128,000 tokens |
| Input | text, image | text, image |
| Output | text | text |
When to choose each
GPT-6 Luna
Best for
- High-volume summarisation, extraction and routing
- Long-context budget work, where the 272k threshold is more forgiving than Haiku 5.5's 100k
Watch out for
- Prompts over 272,000 tokens are rebilled at $0.20 input and $0.75 output for the whole request
- A twentieth of GPT-6 Sol's price reflects a capability gap, not a discount on the same model
GPT-6 Sol
Best for
- General production work that does not need Astra's capability or price
- Workloads kept under the 272,000-token threshold
Watch out for
- Prompts over 272,000 input tokens are rebilled for the whole request at $4 input and $15 output, so crossing the line roughly doubles input cost
- GPT-6.1 Sol lists the same base rates with half the cached-input price, so there is little reason to pick this one for a cache-heavy workload
GPT-6 Luna vs GPT-6 Sol FAQ
Which is cheaper, GPT-6 Luna or GPT-6 Sol?
GPT-6 Luna is 95% cheaper per 1M input tokens than GPT-6 Sol, $0.10 vs $2. For 10M input and 2M output tokens, GPT-6 Luna costs $2 and GPT-6 Sol $40 at standard rates.
Which scores higher on independent benchmarks, GPT-6 Luna or GPT-6 Sol?
GPT-6 Sol leads on Arena Text, 1456 Elo to 1443 Elo; GPT-6 Sol leads on GPQA Diamond, 94.3% to 90.5%; GPT-6 Sol leads on FrontierMath Tiers 1-3 (v2), 89.8% to 78.9%. Scores are independent evaluations, not either vendor's launch figures.
Which has the larger context window, GPT-6 Luna or GPT-6 Sol?
Both accept 1,050,000 tokens of context.
Related comparisons
Production evaluation
Choosing between GPT-6 Luna and GPT-6 Sol?
Public benchmarks narrow the shortlist. We run both on your own tasks, data and cost limits, then build the routing and review controls to run the winner in production.
- A head-to-head evaluation on your real workload
- Routing between models by cost and difficulty