GPT-5.4 vs GPT-5.6 Luna
何時用哪一個
兩者上下文同為 1050000 token,單次最多輸出 128000 token,所以決定因素是價格與工具能力。gpt-5.6-luna 在每一檔費率上都更低:每百萬輸入 $1 對 $2.5,輸出 $6 對 $15,快取讀取 $0.1 對 $1.25 —— 在重複系統提示形成的熱前綴上便宜約 12 倍。需要 computer use 或並行工具呼叫時選 gpt-5.4,這兩項 gpt-5.6-luna 未列出;只看費率的話它沒有任何優勢。
Benchmark 成績
雙方都被測過的 14 項。
廠商公布: Alibaba (Qwen) Anthropic ByteDance DeepSeek Google MiniMax Moonshot OpenAI Tencent Z.ai
定價
| GPT-5.4 | GPT-5.6 Luna | Δ | |
|---|---|---|---|
| 輸入 / 1M tokens | $2.5 | $1 | 2.5× |
| 輸出 / 1M tokens | $15 | $6 | 2.5× |
| 快取讀取 / 1M tokens | $1.25 | $0.1 | 13× |
| 快取寫入 | 不額外計費 | 不額外計費 | - |
費率取自建置時的即時目錄；各模型頁面皆附有目前的費率卡。
它們的相對位置 — 在此計費單位下，所有 64 個 聊天 模型的 每 1M tokens 的輸入價格（對數尺度）
能力
| GPT-5.4 | GPT-5.6 Luna | |
|---|---|---|
| 工具使用 | 是 | 是 |
| 思考控制 | 可配置 | 可配置 |
| 結構化輸出 | 是 | 是 |
| 提示快取 | 隱式（自動） | 隱式（自動） |
| 快取生命週期 | 5-10m, up to 1h | 5-10m, up to 1h |
| 最小快取前綴 | 1024 個 token | 1024 個 token |
規格
| GPT-5.4 | GPT-5.6 Luna | |
|---|---|---|
| 輸入模態 | 文字 影像 | 文字 影像 |
| 輸出模態 | 文字 | 文字 |
| 發布日期 | 2026-03-05 | 2026-07-09 |
| 知識截止日期 | 2025-08 | 2026-02 |
| 上下文視窗 | 1.1M | 1.1M |
| 最大輸出 | 128K | 128K |
| 思考參數 | reasoning.effort | reasoning.effort |
| 可接受的值 | reasoning.effort
| reasoning.effort
|
| 預設值 | none | medium |
規格摘錄自各供應商的文件；供應商未發布的資料列會直接省略，而非自行推測。 完整來源： GPT-5.4 · GPT-5.6 Luna
單一提示詞，兩款模型 — 經由閘道測量
GPT-5.4 通過 · 3 sentences
Enabling reasoning makes the model generate and evaluate additional internal steps before producing an answer, which increases the total tokens processed and therefore the cost per request. It also tends to increase latency and can trigger more tool calls, retries, or longer context retention, all of which compound usage-based charges. So even if user traffic stayed flat, the average cost per interaction rose materially once reasoning was turned on.
輸出 82 tok 延遲 1.9 s
GPT-5.6 Luna 通過 · 3 sentences
Enabling reasoning causes the model to use substantially more computational steps and generated tokens per request, even when the visible answer is similar. Because our provider charges based largely on token consumption and compute, that increased processing translated into roughly twice the cost. We can control the expense by reserving reasoning for complex tasks, setting token and effort limits, and monitoring cost per successful outcome.
輸出 146 tok (+60 思考) 延遲 3.1 s
指令遵循（恰好三句，可數）、受眾適配（面向 CFO 的語氣），以及下方 token 計量所暴露的隱藏思考計費缺口。
GPT-5.4 通過 · 8/8 cases
```python def longest_run(xs): if not xs: return 0 best = 1 cur = 1 for i in range(1, len(xs)): if xs[i] == xs[i - 1]: cur += 1 else: best = max(best, cur) cur = 1 best = max(best, cur) return best ``` The bug is that the original code was counting *matching transitions* between adjacent elements instead of the *length of the run itself*, so a single element started at `0` instead of `1`, making all run lengths off by one. It also only updated `best` inside the equality case, which means the final run could be lost if the loop ended before a change triggered any correction logic. […]
輸出 243 tok 延遲 3.0 s
GPT-5.6 Luna 通過 · 8/8 cases
```python def longest_run(xs): if not xs: return 0 best = cur = 1 for i in range(1, len(xs)): if xs[i] == xs[i - 1]: cur += 1 else: best = max(best, cur) cur = 1 return max(best, cur) ``` The bug is that `cur` counts equal *transitions* rather than the number of items in the current run, so a single-element run returns `0` and a run of length `n` returns `n - 1`; additionally, `best` is only updated when a run is interrupted, so a run continuing through the final element is not finalized before returning. Initializing the current run to one item, resetting it to one after a change, and taking a final `max` fixes both issues.
輸出 356 tok (+161 思考) 延遲 4.7 s
修復是否真的正確（可執行）、解釋的資訊密度，以及在一個邊界明確的任務上的 token 效率。
GPT-5.4 通過 · 5/5 fields, guidance "withheld pending the Q4 close"
{"company":"Meridian Robotics","quarter":"September quarter","revenue_usd_m":412,"growth_pct":11,"guidance":"withheld pending the Q4 close"}
輸出 39 tok 延遲 1.3 s
GPT-5.6 Luna 通過 · 5/5 fields, guidance "Withheld pending the Q4 close"
{"company":"Meridian Robotics","quarter":"September quarter","revenue_usd_m":412,"growth_pct":null,"guidance":"Withheld pending the Q4 close"}
輸出 227 tok (+181 思考) 延遲 3.6 s
schema 服從度（不臆造欄位）、幻覺壓力（guidance 明確被暫緩給出），以及結構化輸出路徑的差異。
GPT-5.4 通過 · 126 words, 0 banned, 1 question
Today we’re introducing Prompt Cache, a new API gateway feature that stores prompt results and reuses them across supported model providers. Teams can reduce latency, lower repeated compute costs, and keep application behavior more consistent when identical or near-identical prompts appear in production. Prompt Cache works at the gateway layer, so developers can adopt it without rewriting provider integrations or changing routing logic. Built-in controls let you set cache keys, expiration policies, and privacy rules for sensitive workloads. Need faster responses without duplicating optimization work for each vendor? With shared observability, you can track hit rates, savings, and cache performance in one place. […]
輸出 154 tok 延遲 2.9 s
GPT-5.6 Luna 通過 · 120 words, 0 banned, 1 question
Introducing PromptCache, an API gateway feature that caches prompts across providers, helping teams reduce latency, control spend, and deliver consistent results. How much faster could your applications respond when repeated prompts are served from a shared cache instead of being sent upstream? PromptCache supports provider-aware routing, configurable time-to-live policies, encrypted storage, cache invalidation, and usage analytics through one operational layer. It works with language-model providers while preserving your authentication, observability, and fallback workflows. Developers can enable caching by endpoint, model, tenant, or prompt pattern, then monitor hit rates and savings in real time. Built for production workloads, PromptCache gives platform teams controls for performance and cost without requiring application rewrites. […]
輸出 948 tok (+778 思考) 延遲 8.0 s
約束服從度（字數預算、禁用詞表、唯一的那句問句）、文風指紋，以及長度控制。
只需一行程式碼即可在兩者間切換
以下每個頁籤中都有這兩個 ID — 醒目提示的這兩行是唯一的修改處。相同的端點，相同的金鑰，相同的請求結構。
from openai import OpenAI
client = OpenAI(
base_url="https://synthorai.io/v1",
api_key="sk-syn-...",
)
resp = client.chat.completions.create(
model="gpt-5.4",
# model="gpt-5.6-luna", # 取消註解此行，並註解上一行
messages=[{"role": "user", "content": "Summarize this diff"}],
reasoning_effort="medium",
)
print(resp.choices[0].message.content)import OpenAI from "openai";
const client = new OpenAI({
baseURL: "https://synthorai.io/v1",
apiKey: "sk-syn-...",
});
const resp = await client.chat.completions.create({
model: "gpt-5.4",
// model: "gpt-5.6-luna", // 取消註解此行，並註解上一行
messages: [{ role: "user", content: "Summarize this diff" }],
reasoning_effort: "medium",
});
console.log(resp.choices[0].message.content);curl https://synthorai.io/v1/chat/completions \
-H "Authorization: Bearer sk-syn-..." \
-H "Content-Type: application/json" \
-d '{
"model": "gpt-5.4",
# "model": "gpt-5.6-luna", # 取消註解此行，並註解上一行
"messages": [{"role": "user", "content": "Hello"}],
"reasoning_effort": "medium"
}'package main
import (
"context"
"fmt"
"github.com/openai/openai-go/v3"
"github.com/openai/openai-go/v3/option"
)
func main() {
client := openai.NewClient(
option.WithBaseURL("https://synthorai.io/v1"),
option.WithAPIKey("sk-syn-..."),
)
resp, _ := client.Chat.Completions.New(context.TODO(), openai.ChatCompletionNewParams{
Model: "gpt-5.4",
// Model: "gpt-5.6-luna", // 取消註解此行，並註解上一行
Messages: []openai.ChatCompletionMessageParamUnion{
openai.UserMessage("Summarize this diff"),
},
ReasoningEffort: openai.ReasoningEffortMedium,
})
fmt.Println(resp.Choices[0].Message.Content)
}import com.openai.client.OpenAIClient;
import com.openai.client.okhttp.OpenAIOkHttpClient;
import com.openai.models.chat.completions.*;
import com.openai.models.ReasoningEffort;
OpenAIClient client = OpenAIOkHttpClient.builder()
.baseUrl("https://synthorai.io/v1")
.apiKey("sk-syn-...")
.build();
ChatCompletion resp = client.chat().completions().create(
ChatCompletionCreateParams.builder()
.model("gpt-5.4")
// .model("gpt-5.6-luna") // 取消註解此行，並註解上一行
.addUserMessage("Summarize this diff")
.reasoningEffort(ReasoningEffort.MEDIUM)
.build());
System.out.println(resp.choices().get(0).message().content().orElse(""));常見問題
GPT-5.4 和 GPT-5.6 Luna 哪個比較便宜？
GPT-5.6 Luna 在 輸入 / 1m tokens 上較便宜（$1 對比 $2.5，相差 2.5×）。其他項目可能呈現相反結果 — 上表提供完整資訊，實際成本取決於您的使用組合。
我可以在不進行兩次整合的情況下，對 GPT-5.4 和 GPT-5.6 Luna 進行 A/B 測試嗎？
可以。兩者皆透過同一個相容 OpenAI 的端點提供服務，並使用同一把 API 金鑰 — 切換只需更改一行的模型字串，因此您可以將部分流量分別導向兩者並直接比較帳單。
GPT-5.4 與 GPT-5.6 Luna 支援提示快取嗎？
是的 — 兩者的快取讀取費率皆低於其輸入費率，因此具有暖前綴的工作負載成本會低於牌價所示。確切的快取讀取列請見上方的定價表。