GPT-6 Luna API

Q: Which model ID should I use?

Use gpt-6-luna. There is no generic gpt-6 alias, and the same ID works on Chat Completions and Responses.

Q: How much does GPT-6 Luna cost on EvoLink?

EvoLink default rates are $0.092 input and $0.450 output per 1M tokens, against OpenAI’s $0.100 / $0.500 list price. Cache reads are $0.0089 and cache writes $0.114 per 1M tokens; logged-in users see the rates for their own account.

Q: How is cached input billed?

Cache reads and cache writes are separate token segments. Neither segment is billed again as ordinary input.

Q: When do long-context rates apply?

When total prompt tokens exceed 272,000, the whole request uses the long-context multipliers: 2x for input and both cache roles, 1.5x for output.

Q: Which reasoning effort levels are supported?

reasoning_effort on Chat Completions and reasoning.effort on Responses accept none, low, medium (the default), high, xhigh, and max. none skips reasoning for the lowest latency and cost; higher levels spend reasoning tokens, which are billed as output.

Q: Can I use function calling on Chat Completions?

Only with reasoning_effort set to none. For built-in tools, or function calling at a higher reasoning effort, call the Responses API with the same gpt-6-luna ID.

Q: What are the context window and maximum output?

GPT-6 Luna has a 1,050,000-token context window, with up to 922,000 input tokens and up to 128,000 output tokens per request.

Q: Should I migrate my queue from GPT-5.6 Luna to GPT-6 Luna?

GPT-6 Luna keeps the 1.05M context, 128K output and six reasoning levels of GPT-5.6 Luna, at half the OpenAI list price, with a newer knowledge cutoff (May 18, 2026). OpenAI reports that at higher effort it matches GPT-5.6 Sol's factuality at about one-hundredth of the task cost, with a deception rate of 2.8% versus 9.5% for GPT-5.6 Luna; those are vendor numbers. Replay your own records and decide on cost per accepted record, not price per token; the GPT-6 Luna vs GPT-5.6 Luna guide below has the field-level acceptance set and the one contract change to check first: function calling on Chat Completions requires reasoning_effort none.

Q: GPT-6 Luna, Sol or Astra: which one should I use?

Use Luna for focused, high-volume calls such as classification, extraction, routing and subagents; move complex coding and tool-driven agents to GPT-6 Sol; send the hardest end-to-end reasoning and research to GPT-6 Astra. All three share the 1.05M context, the same integration and the same 272K long-context rule, so you can split traffic by task and keep one API key. The GPT collection page compares them side by side.

Access OpenAI GPT-6 Luna—also searched as GPT 6 Luna / gpt-6-luna— through EvoLink's unified chat API. Test low-latency routing, structured output, tool use before integrating.

OpenAIText GenerationAvailable

from $0.092 / 1M input tokens$0.100 official price-9%

API docs

Chat CompletionsResponses APIReasoning effortPrompt cachingStructured outputs

Production routeLive

Live

Context

1.05M context · 128K max output

Best For

High-volume classification, extraction, routing, subagents

Input

Text + images

Output

Text · JSON (structured output) · tool calls

Models Pricing Readme API

Choose GPT-6 Luna

One exact model ID on the OpenAI-compatible integration. Short- and long-context rates for all four token roles are listed on the card.

GPT-6 Luna

OpenAI's most efficient GPT-6 model for focused, high-volume tasks.

Selected

Model ID

gpt-6-luna

Best for

High-volume classification, extraction, routing, and subagent calls where latency and cost matter most

< 272K

Input

$0.092 / 1M-9%

6.2 cr / 1M$0.100official price

Cache read

$0.0089 / 1M-12%

0.6 cr / 1M$0.010official price

Cache write

$0.114 / 1M-9%

7.7 cr / 1M$0.125official price

Output

$0.450 / 1M-10%

30.6 cr / 1M$0.500official price

≥ 272K

Input

$0.183 / 1M-9%

12.4 cr / 1M$0.200official price

Cache read

$0.018 / 1M-12%

1.2 cr / 1M$0.020official price

Cache write

$0.227 / 1M-9%

15.4 cr / 1M$0.250official price

Output

$0.675 / 1M-10%

45.9 cr / 1M$0.750official price

All rates are per 1M tokens, shown in USD and credits, and reflect your account's current pricing. Above 272K prompt tokens the long-context tier applies: ×2 for input and cache roles, ×1.5 for output.

GPT-6 Luna pricing

Estimate what one GPT-6 Luna request costs before you integrate. The calculator uses your account's current rates, with official pricing as a reference.

Request calculator

Enter the token mix for one request.

Input tokensCache read tokensCache write tokensOutput tokens

Estimated request cost

GPT-6 Luna

Short-context rate

USD$0.0003

Credits0.0156

Official estimate $0.0003 · save $0.0001 (9%)

Input tokens0.0062 cr

Cache read tokens0.0002 cr

Cache write tokens0 cr

Output tokens0.0092 cr

Minimum charge: 0.01 credits per request. At 272K prompt tokens, input and cache roles are billed at ×2 and output at ×1.5. Tool calls are not multiplied.

Budget guide

Approximate requests using the current token mix.

Add credits

$10

About 41666 requests

For quick testing

$50

About 217948 requests

For regular development

$100

About 435897 requests

For production evaluation

GPT-6 Luna API

OpenAI's most efficient GPT-6 model for focused, high-volume tasks — classification, extraction, routing, and subagents — available through one EvoLink API key with Chat Completions and Responses.

GPT-6 Luna is served on EvoLink under the model ID gpt-6-luna through Chat Completions · Responses, with the same API key and balance you use for every other model. It offers a 1.05M context window and up to 128K output tokens, plus low-latency routing, structured output, tool use.

GPT-6 Luna specs and capabilities

Numbers come from the EvoLink route configuration; capabilities are what the API exposes today.

Context window: 1.05M tokens
Max output: 128K tokens
Input: Text + images
Output: Text · JSON (structured output) · tool calls
Reasoning: Configurable reasoning effort
Tool use: Function calling with multi-step tool sequences
Prompt caching: Cache write + cache read rates
Long-context tier: ×2 in / ×1.5 out above 272K prompt tokens
Protocols: Chat Completions · Responses
Model ID: gpt-6-luna

What GPT-6 Luna is best for

One exact model ID for focused, repeatable calls at volume; hand open-ended coding and agent work to GPT-6 Sol or GPT-6 Astra. OpenAI reports that at higher effort it matches GPT-5.6 Sol's factuality at about one-hundredth of the task cost, with a 2.8% deception rate versus 9.5% for GPT-5.6 Luna (vendor-reported).

Classify and Route at Volume

Label tickets, intents, and documents with short prompts; reasoning_effort none keeps latency and output tokens low.

Extract Structured Data

Turn emails, forms, and logs into JSON with structured outputs, and validate the schema before writing to your systems.

Run Subagents Under a Larger Model

Hand focused sub-tasks from GPT-6 Sol or GPT-6 Astra to Luna, and keep planning and orchestration on the stronger model.

Reuse Long Prompts with Caching

Keep stable instructions and examples at the start of the prompt so repeated calls are billed as cache reads instead of full-price input.

Two ways to use GPT-6 Luna: EvoLink API or Agent

Use the EvoLink API for product backends and batch jobs, or call GPT-6 Luna from Codex, Claude, or Gemini for coding and analysis workflows. Both paths share the same EvoLink API key, balance, model ID, and request history.

Option 1

Integrate with the EvoLink API

Best for: product backends, batch jobs, automated pipelines

Send OpenAI-compatible Chat Completions (or Anthropic Messages) requests to EvoLink and control the model ID, system prompt, output budget, tools, and structured output.

1Create an EvoLink API key in the console
2Point your OpenAI or Anthropic SDK at the EvoLink base URL and select the model ID shown above
3Send one representative request and read the usage field for input, cached, and output tokens
4Set max_tokens and retries per task; keep tool-call IDs and results across turns

Create EvoLink API Key

Option 2

Call it with an Agent

Best for: coding, review, and analysis tasks in Codex, Claude, and Gemini

Give the Agent the task, the inputs to include, and the acceptance criteria. It assembles the request, calls GPT-6 Luna through EvoLink, and returns the answer with token usage.

1Set EVOLINK_API_KEY in your local environment; never put it in code or a prompt
2Describe the task, the inputs to include, and the expected output format
3Ask the Agent to call GPT-6 Luna through EvoLink and show the request before sending
4Let the Agent report the answer, token usage, and any error body

GPT-6 Luna API code example and error handling

This example shows the shortest runnable request: an OpenAI-compatible Chat Completions call with a system prompt, a user message, and an output budget. Open the API tab for the complete parameter and response reference.

View complete API docs

cURL

curl -X POST https://api.evolink.ai/v1/chat/completions \
  -H "Authorization: Bearer $EVOLINK_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "gpt-6-luna",
    "messages": [
      { "role": "system", "content": "Classify the support ticket as billing, bug, or how-to. Reply with JSON only." },
      { "role": "user", "content": "My invoice shows two charges for the same month." }
    ],
    "reasoning_effort": "none",
    "response_format": { "type": "json_object" },
    "max_tokens": 200
  }'

# Use the exact ID gpt-6-luna on both Chat Completions and Responses.
# reasoning_effort accepts none, low, medium (default), high, xhigh, and
# max; none skips reasoning for the lowest latency and cost, and reasoning
# tokens at other levels are billed as output. On Chat Completions, function
# calling works only with reasoning_effort set to none.

Invalid request or unsupported parameter

Check the model ID, the messages array, and parameter ranges against the API reference; remove fields this route does not support.

Authentication or balance issue

Check the Authorization bearer token and confirm the available balance in the console.

Context length exceeded

Prompt tokens exceed the GPT-6 Luna context window. Trim or retrieve only the relevant evidence and reuse cached prefixes.

Rate limited (429)

Back off and retry with jitter; batch or queue requests instead of sending parallel bursts.

Content or tool call rejected

Review sensitive content, malformed tool-call arguments, and JSON schema mismatches before retrying.

Before you take GPT-6 Luna to production

Four checks that decide cost and stability.

Pick the exact GPT-6 Luna model ID

Use gpt-6-luna on both Chat Completions and Responses; there is no generic gpt-6 alias.

Required

Start at reasoning_effort none

Begin at none for classification and extraction, and raise effort only where evaluation shows a quality gain; reasoning tokens are billed as output.

Recommended

Choose the endpoint for tool calls

Use Responses for built-in tools and function calling; on Chat Completions, function calling works only with reasoning_effort none.

Required

Budget for the 272K threshold and measure with usage

Above 272K prompt tokens the whole request uses long-context rates; reconcile cost from the usage object before scaling traffic.

Recommended

Four-part prompt caching

Uncached input, cache reads, cache writes, and output are settled separately from the final usage object. Cache writes are not charged again as ordinary input.

Long-context threshold

At more than 272K prompt tokens, input and both cache roles use 2x rates while output uses 1.5x rates for the entire request.

GPT-6 Luna vs GPT-6 Sol and GPT-5.6 Luna

EvoLink

Compare GPT-6 Luna with GPT-6 Sol and GPT-5.6 Luna using EvoLink prices per 1M tokens.

Feature	GPT-6 Luna	GPT-6 Sol	GPT-5.6 Luna
EvoLink price / 1M (input / output)	$0.092 / $0.45	$1.8 / $9	$0.181 / $1.081
Context window	1.05M	1.05M	1.05M
Prompt caching	Cache read + cache write	Cache read + cache write	Cache read + cache write
Best for	High-volume classification, extraction, routing, and subagent calls where latency and cost matter most	Complex coding and agentic workflows, one tier up	Previous-generation Luna tier for high-volume, cost-sensitive routes

GPT-6 Luna

EvoLink price / 1M (input / output)$0.092 / $0.45

Context window1.05M

Prompt cachingCache read + cache write

Best forHigh-volume classification, extraction, routing, and subagent calls where latency and cost matter most

GPT-6 Sol

EvoLink price / 1M (input / output)$1.8 / $9

Context window1.05M

Prompt cachingCache read + cache write

Best forComplex coding and agentic workflows, one tier up

GPT-5.6 Luna

EvoLink price / 1M (input / output)$0.181 / $1.081

Context window1.05M

Prompt cachingCache read + cache write

Best forPrevious-generation Luna tier for high-volume, cost-sensitive routes

Base-context prices reflect your account's current pricing; signed-out visitors see Default rates. Long-context multipliers are listed in the Pricing section.

GPT Model Family

Same API key and balance — switch tiers without changing your integration.

Compare all GPT models

GPT-6.1 Sol

Updated Sol for multi-file coding and Responses-based agents

View model

GPT-6 Astra

OpenAI's most capable model for the hardest end-to-end work

View model

GPT-6 Sol

GPT-6 model built for complex coding and agentic workflows

View model

Other text models on EvoLink besides GPT-6 Luna

Gemini 3.5 Flash-Lite

Google's lowest-cost Flash Lite model with multimodal input, built for high-volume, latency-sensitive, and batch workloads.

View model

DeepSeek V4.1 Flash

DeepSeek Flash route with text and image input, a 1M context window, and configurable thinking.

View model

GLM-5.3 Flash

Z.ai's multimodal volume tier of the GLM-5.3 generation, with a 1M-token context window.

View model

Grok 4.5

xAI reasoning and tool-use model with a 500K context window and metered server-side tools.

View model

GPT-6 Luna guides and related reading

GPT-6 Luna release date and rollout

Released September 22, 2026; rollout by channel and what OpenAI reports for factuality and cost per task

Read guide

GPT-6 Luna vs GPT-5.6 Luna

Half the list price, fewer reported mistakes, the Chat Completions tool-calling rule, and a cost-per-accepted-record migration gate

Read guide

GPT-6 API guide (Astra)

First call, reasoning effort, the 272K and cache billing rules shared across the GPT-6 family

Read guide

GPT-6 Luna API - FAQ

Which model ID should I use?

Use gpt-6-luna. There is no generic gpt-6 alias, and the same ID works on Chat Completions and Responses.

How much does GPT-6 Luna cost on EvoLink?

EvoLink default rates are $0.092 input and $0.450 output per 1M tokens, against OpenAI’s $0.100 / $0.500 list price. Cache reads are $0.0089 and cache writes $0.114 per 1M tokens; logged-in users see the rates for their own account.

How is cached input billed?

Cache reads and cache writes are separate token segments. Neither segment is billed again as ordinary input.

When do long-context rates apply?

When total prompt tokens exceed 272,000, the whole request uses the long-context multipliers: 2x for input and both cache roles, 1.5x for output.

Which reasoning effort levels are supported?

reasoning_effort on Chat Completions and reasoning.effort on Responses accept none, low, medium (the default), high, xhigh, and max. none skips reasoning for the lowest latency and cost; higher levels spend reasoning tokens, which are billed as output.

Can I use function calling on Chat Completions?

Only with reasoning_effort set to none. For built-in tools, or function calling at a higher reasoning effort, call the Responses API with the same gpt-6-luna ID.

What are the context window and maximum output?

GPT-6 Luna has a 1,050,000-token context window, with up to 922,000 input tokens and up to 128,000 output tokens per request.

Should I migrate my queue from GPT-5.6 Luna to GPT-6 Luna?

GPT-6 Luna keeps the 1.05M context, 128K output and six reasoning levels of GPT-5.6 Luna, at half the OpenAI list price, with a newer knowledge cutoff (May 18, 2026). OpenAI reports that at higher effort it matches GPT-5.6 Sol's factuality at about one-hundredth of the task cost, with a deception rate of 2.8% versus 9.5% for GPT-5.6 Luna; those are vendor numbers. Replay your own records and decide on cost per accepted record, not price per token; the GPT-6 Luna vs GPT-5.6 Luna guide below has the field-level acceptance set and the one contract change to check first: function calling on Chat Completions requires reasoning_effort none.

GPT-6 Luna, Sol or Astra: which one should I use?

Use Luna for focused, high-volume calls such as classification, extraction, routing and subagents; move complex coding and tool-driven agents to GPT-6 Sol; send the hardest end-to-end reasoning and research to GPT-6 Astra. All three share the 1.05M context, the same integration and the same 272K long-context rule, so you can split traffic by task and keep one API key. The GPT collection page compares them side by side.