DeepSeek Flash (V4.1) vs GPT-6 Luna.
GPT-6 Luna is 2.6x cheaper.
For a chatbot doing 100K requests a month with 1,000 input and 500 output tokens each: $35.00 vs $90.00.
DeepSeek
DeepSeek Flash (V4.1)
| Input / 1M | $0.30 |
| Cached / 1M | $0.006 |
| Output / 1M | $1.20 |
| Context | 1M |
| Chatbot at 100K req/mo | $90.00 |
Source · Oct 20, 2026
OpenAI
GPT-6 Luna
| Input / 1M | $0.10 |
| Cached / 1M | $0.01 |
| Output / 1M | $0.50 |
| Context | — |
| Chatbot at 100K req/mo | $35.00 |
Source · Oct 20, 2026
Sponsored
Your API or dev tool here. Reach developers comparing model costs right now.Price isn't everything: check quality on your own prompts. GPT-6 Luna wins on list price for this workload, but output-heavy jobs (long answers, agents, reasoning) shift the math toward whichever model has the lower output price, which here is GPT-6 Luna.