GPT-6 Luna vs Gemini 3.5 Flash-Lite
GPT-6 Luna costs $0.10 / $0.50 and Gemini 3.5 Flash-Lite costs $0.30 / $2.50 per million input / output tokens. At a typical 3:1 mix, GPT-6 Luna is 4.3× cheaper than Gemini 3.5 Flash-Lite. GPT-6 Luna also scores clearly higher on Epoch AI’s Capabilities Index (157 vs 146).
Price
GPT-6 Luna is 4.3× cheaper per token
On a coding agent: $25.98 vs $84.46 a month (3× less)
Capability
GPT-6 Luna scores higher
GPT-6 Luna 157 (155–159)
Gemini 3.5 Flash-Lite 146 (144–147)
Gemini 3.5 Flash-Lite 146 (144–147)
Best value at GPT-6 Luna’s level
GPT-6 Luna
The cheapest current model that scores level or higher, on a coding agent
Input
$0.10
per 1M
Output
$0.50
per 1M
Cached
$0.010
1.05M context
Input
$0.30
per 1M
Output
$2.50
per 1M
Cached
$0.030
1M context
Side by side
✓ marks the cheaper price, larger context window, bigger discount or higher score.
| Detail | GPT-6 Luna | Gemini 3.5 Flash-Lite |
|---|---|---|
| Price per 1M tokens | ||
| Input | $0.10Lowest | $0.30 |
| Output | $0.50Lowest | $2.50 |
| Blended (3 in : 1 out) | $0.20Lowest | $0.85 |
| Cached input | $0.010Lowest | $0.030 |
| Batch or off-peak discount | 50% (batch) | 50% (batch) |
| Long prompts | Over 272K: 2× in, 1.5× out | Same rate at any length |
| Limits and model | ||
| Context window | 1.05MLargest | 1M |
| Capability score Epoch AI index | 157Highest | 146 |
| Released | 22 Sep 2026 | 21 Jul 2026 |
| Open weights | No | No |
| Status | Current | Current |
| Checked | ||
| Price checked | 2 Nov 2026 OpenAI pricing | 6 Nov 2026 Gemini API pricing |
Monthly cost on three workloads
List prices, no batch discount. The cached share applies where a model has a cache rate.
| Workload | GPT-6 Luna | Gemini 3.5 Flash-Lite | Difference |
|---|---|---|---|
| Customer chat assistant50,000 conversations a month, about 2,000 tokens in and 400 out each; 50% cachedPrice this workload in the calculator | $15.50Lower | $66.50 | $51.00 |
| Document Q&A (RAG)20,000 questions a month over retrieved documents, 12,000 in and 600 out; 30% cachedPrice this workload in the calculator | $23.52Lower | $82.56 | $59.04 |
| Coding agent400 tasks a month, 40 steps each with a growing context; 85% cachedPrice this workload in the calculator | $21.63Lower | $72.88 | $51.25 |
How they differ
- GPT-6 Luna is cheaper on input ($0.10 vs $0.30).
- GPT-6 Luna is cheaper on output ($0.50 vs $2.50), which matters most for long answers and reasoning models.
- GPT-6 Luna accepts longer prompts (1.05M vs 1M tokens).
- Cached input is cheaper on GPT-6 Luna ($0.01 vs $0.03), which favours agents and chat apps that resend the same context.
- On Epoch AI’s Capabilities Index, GPT-6 Luna scores clearly higher (157 vs 146), beyond the margin of error.
Recent changes
All changes- OpenAI launched GPT-6.1 Terra at $1 / $5 per 1M tokens, with cached input at $0.05: half the price of GPT-6.1 Sol.
- GPT-6.1 Sol replaced GPT-6 Sol at the same $2 / $10 per 1M tokens, with cached input cut from $0.20 to $0.10.
- OpenAI launched GPT-6 Sol at $2 / $10 and GPT-6 Luna at $0.10 / $0.50.
- OpenAI launched GPT-6 Astra at $10 / $50, its most expensive standard model.
- ▼ 50%Google launched Gemini 3.6 Flash at $0.75 / $3.75, half the price of 3.5 Flash, and Gemini 3.5 Flash-Lite at $0.30 / $2.50.
$1.50 / $9→ $0.75 / $3.75
