GPT-6 Luna vs DeepSeek V4.1 Flash
GPT-6 Luna is the lower-priced of the two on Router One at base-tier rates (GPT-6 Luna: request input ≤ 272,000 tokens) — about 80% less on a 1M-input + 1M-output mix. GPT-6 Luna offers 1.05M of context vs 1M for DeepSeek V4.1 Flash. GPT-6 Luna answers on /v1/chat/completions and /v1/responses (Codex CLI); DeepSeek V4.1 Flash on /v1/chat/completions, /v1/messages (Claude Code) and /v1/responses (Codex CLI). GPT-6 Luna is in no plan and bills the wallet; DeepSeek V4.1 Flash counts against the Standard models allowance on Pro, Max and Ultra.
GPT-6 Luna and DeepSeek V4.1 Flash compared on current per-token rates, context window, and capabilities — both callable through one OpenAI-compatible endpoint with per-request cost traces.
GPT-6 Luna vs DeepSeek V4.1 Flash: rates, context window, and capabilities
| Spec | GPT-6 Luna | DeepSeek V4.1 Flash |
|---|---|---|
| Input / 1M tokens | $0.10 | $0.60 |
| Output / 1M tokens | $0.50 | $2.40 |
| Cached input / 1M tokens | $0.01 | $0.012 |
| Context window | 1.05M | 1M |
| Capabilities | Chat, Streaming, Tool calling, Vision | Chat, Streaming, Tool calling, Vision |
| Subscription plans | Not in any plan — wallet | Standard models — Pro, Max, Ultra |
| Detail page | GPT-6 Luna | DeepSeek V4.1 Flash |
GPT-6 Luna: request input ≤ 272,000 tokens: Input $0.10 / 1M tokens, Output $0.50 / 1M tokens, Cache write $0.125 / 1M tokens, Cached input $0.01 / 1M tokens; request input > 272,000 tokens: Input $0.20 / 1M tokens, Output $0.75 / 1M tokens, Cache write $0.25 / 1M tokens, Cached input $0.02 / 1M tokens. The total input tokens in each request select the tier; its rates apply to the whole request, not only tokens above the threshold.
What does 1M tokens cost on GPT-6 Luna vs DeepSeek V4.1 Flash?
Base-tier rates: For a workload of 1M input plus 1M output tokens at current rates: GPT-6 Luna comes to $0.60, DeepSeek V4.1 Flash comes to $3.00 — GPT-6 Luna is about 80% cheaper on this mix. Real workloads skew heavily toward input tokens, so weigh the input rate by your own ratio; the cached-input row above is the posted catalog rate for that model.
Switch between GPT-6 Luna and DeepSeek V4.1 Flash
The API key and base URL stay the same, but this is not a one-string switch for every client or request. The model pages list GPT-6 Luna on /v1/responses and /v1/chat/completions, and Router One serves DeepSeek V4.1 Flash on /v1/chat/completions, /v1/messages and /v1/responses, so Claude Code (/v1/messages) runs only the DeepSeek side. Per OpenAI's GPT-6 Luna model page and GPT-6 guide (checked 2026-10-31), GPT-6 Luna takes built-in tools and function calls on the Responses API, supports function calling on Chat Completions only with reasoning_effort none, and at any reasoning effort other than none takes no temperature, top_p or logprobs settings, so a DeepSeek V4.1 Flash request that sends tools on /v1/chat/completions or sets those sampling parameters has to change before it runs on GPT-6 Luna. Per DeepSeek's thinking mode guide (checked 2026-10-31), DeepSeek V4.1 Flash runs in thinking mode by default at high effort and treats a requested medium or xhigh effort as high; in thinking mode, setting temperature, presence_penalty or frequency_penalty returns no error but has no effect. On a Pro, Max or Ultra plan the billing differs too: as of the 2026-10-31 plan response, DeepSeek V4.1 Flash counts against the Standard models allowance, while GPT-6 Luna, in no plan tier, bills the wallet per token.
curl https://api.router.one/v1/chat/completions \
-H "Authorization: Bearer sk-your-router-one-key" \
-H "Content-Type: application/json" \
-d '{"model": "openai/gpt-6-luna", "messages": [{"role": "user", "content": "Hello"}]}'
# Same request, other model — change one string:
# "model": "deepseek-v4.1-flash"FAQ
Is GPT-6 Luna cheaper than DeepSeek V4.1 Flash?
Compared at base-tier rates (GPT-6 Luna: request input ≤ 272,000 tokens). Input: GPT-6 Luna $0.10 vs DeepSeek V4.1 Flash $0.60 / 1M tokens; GPT-6 Luna has the lower rate. Output: GPT-6 Luna $0.50 vs DeepSeek V4.1 Flash $2.40 / 1M tokens; GPT-6 Luna has the lower rate. Total cost depends on the workload's input, output and cache usage. GPT-6 Luna: request input ≤ 272,000 tokens: Input $0.10 / 1M tokens, Output $0.50 / 1M tokens, Cache write $0.125 / 1M tokens, Cached input $0.01 / 1M tokens; request input > 272,000 tokens: Input $0.20 / 1M tokens, Output $0.75 / 1M tokens, Cache write $0.25 / 1M tokens, Cached input $0.02 / 1M tokens. The total input tokens in each request select the tier; its rates apply to the whole request, not only tokens above the threshold. Rates change; the /models page is the live source of truth.
Is GPT-6 Luna or DeepSeek V4.1 Flash included in a Router One subscription?
GPT-6 Luna is in no Router One plan, so every call bills the wallet per token. DeepSeek V4.1 Flash counts against the Standard models allowance on Pro, Max and Ultra: Pro 5,000, Max 8,000 and Ultra 25,000 requests per 30-day cycle, shared by every model in that tier. Requests beyond a plan's allowance bill the wallet per token.
Can I switch between GPT-6 Luna and DeepSeek V4.1 Flash without changing code?
The API key and base URL stay the same, but this is not a one-string switch for every client or request. The model pages list GPT-6 Luna on /v1/responses and /v1/chat/completions, and Router One serves DeepSeek V4.1 Flash on /v1/chat/completions, /v1/messages and /v1/responses, so Claude Code (/v1/messages) runs only the DeepSeek side. Per OpenAI's GPT-6 Luna model page and GPT-6 guide (checked 2026-10-31), GPT-6 Luna takes built-in tools and function calls on the Responses API, supports function calling on Chat Completions only with reasoning_effort none, and at any reasoning effort other than none takes no temperature, top_p or logprobs settings, so a DeepSeek V4.1 Flash request that sends tools on /v1/chat/completions or sets those sampling parameters has to change before it runs on GPT-6 Luna. Per DeepSeek's thinking mode guide (checked 2026-10-31), DeepSeek V4.1 Flash runs in thinking mode by default at high effort and treats a requested medium or xhigh effort as high; in thinking mode, setting temperature, presence_penalty or frequency_penalty returns no error but has no effect. On a Pro, Max or Ultra plan the billing differs too: as of the 2026-10-31 plan response, DeepSeek V4.1 Flash counts against the Standard models allowance, while GPT-6 Luna, in no plan tier, bills the wallet per token.
Where do these numbers come from?
Specs and prices on this page render from the live Router One catalog — the same data as the /models page — and refresh with it. Pricing methodology is documented on /pricing-methodology. Rates on this page were read from the live catalog on 2026-10-31 (UTC) and refresh within the hour.