跳轉到主要內容

GPT-5.6 Luna (max)

OpenAIGPTProprietary

描述

GPT-5.6 Luna is the fastest and most cost-efficient tier of OpenAI's GPT-5.6 family, designed for cost-sensitive, high-volume workloads. It brings strong capability at OpenAI's lowest GPT-5.6 price, supports max reasoning effort, and has a 1.05M-token context window.

發布日期
2026-07-09
參數規模
上下文長度
1.1M
支援模態
image, pdf, text

能力雷達圖

50
general
68
coding
91
reasoning
67
science
70
agents
85
multimodal

排行榜排名

領域#排名分數來源
智慧體能力模型榜50
52.0
LS
程式碼能力榜20
92.0
AA
通用能力榜34
83.0
AA
多模態榜70
43.0
LS
科學能力38
85.0
AA

基準測試分數 (LLM Stats)

(LLM Stats (zeroeval))

Agents

Connectors99.9%自報
Search and Function-Calling89.7%自報
Capture-the-Flag Challenges (Internal)85.2%自報
Toolathlon53.4%自報
LifeSciBench51.2%自報
Internal Research Debugging Evaluation50.8%自報
Agents' Last Exam50.3%自報
SEC-bench Pro48.9%自報
RSI Index41.9%自報
Big Finance Bench36.0%自報
Management Consulting Tasks (Internal)35.4%自報
PostTrainBench Lite29.6%自報
KernelGen 1P22.4%自報
AutomationBench14.9%自報
ExploitGym12.4%自報
GeneBench-Pro10.8%自報
NanoGPT1.7%自報

Code

BenchCAD (with Python tool)73.9%自報
DeepSWE67.2%自報
DeepSWE 1.167.0%
BenchCAD63.1%自報
ExploitBench33.2%自報

General

Artificial Analysis51.0%
GDP.pdf22.7%自報

Healthcare

HealthBench Consensus95.1%自報
HealthBench55.8%自報
HealthBench Professional55.7%自報
HealthBench Hard32.0%自報

Long Context

MRCR v2 (8-needle, 512K-1M)41.3%自報
MRCR v2 (8-needle)41.3%自報

Math

FrontierMath78.6%自報
FrontierMath Tier 4 (v2)58.5%自報

Multimodal

OSWorld 2.045.6%自報

Reasoning

GPQANYU + Cohere + Anthropic (2023)92.3%自報
Terminal-Bench 2.184.7%自報
BrowseCompOpenAI (2025)83.3%自報
Graphwalks BFS >128k81.3%自報
SWE-Bench ProPrinceton NLP (2024)62.7%自報
Graphwalks BFS 1M51.2%自報
FrontierCode 1.139.8%
MedChemBench (Internal)30.4%自報
ARC-AGI-30.2%自報

Vision

MMMU-Pro (with tools)79.5%自報
MMMU-Pro78.4%自報

AA 評測指數

(Artificial Analysis)
Coding Index(Artificial Analysis)
71.4
Intelligence Index(Artificial Analysis)
52.3
Gpqa(NYU + Cohere + Anthropic (2023))
0.9
Terminalbench V2 1
0.8
Lcr(Artificial Analysis)
0.8
Scicode(UIUC + Argonne National Lab (2024))
0.5
Hle(Center for AI Safety + Scale AI (2025))
0.4
Tau Banking
0.3

LLM Stats 分類評分

(LLM Stats (zeroeval))
Physics
90
Biology
90
Chemistry
90
Search
80
Math
70
Spatial Reasoning
70
Tool Calling
70
Multimodal
60
Vision
60
Long Context
50
Reasoning
50
General
50
Healthcare
50
Agents
50
Code
50
Safety
40
Finance
40
Science
30
Systems
20

定價

輸入價格$0.2 / 1M tokens
輸出價格$1.2 / 1M tokens
混合價格(3:1)$0.45 / 1M tokens
快取讀取價格$0.02 / 1M tokens
快取寫入價格$0.25 / 1M tokens

速度

Tokens/秒132.4
首Token延遲86.96s
首回答延遲86.96s

供應商價格排行

供應商價格排行

2 個供應商

最便宜: OpenAI最貴: Neon
供應商輸入輸出
1OpenAI最便宜
$0
$0
2Neon
$1
$6

比較該模型在不同 API 供應商之間的定價。

外部連結