跳轉到主要內容

Grok 4.6 (high)

SpaceXAIGrokProprietary

描述

Grok 4.6 is SpaceXAI's frontier model built for coding, agentic tasks, and knowledge work. It builds on Grok 4.5 with a particular focus on long-running agents and more ambitious interactive and visual work. The model supports text and image inputs, reasoning, function calling, structured outputs, and a 500K-token context window.

發布日期
2026-08-12
參數規模
—
上下文長度
500K
支援模態
image, pdf, text

能力雷達圖

44
general
74
coding
95
reasoning
71
science
20
agents
80
multimodal

排行榜排名

領域#排名分數來源
程式碼能力榜27
93.0
AA
通用能力榜35
76.0
AA
科學能力37
83.0
AA

基準測試分數 (LLM Stats)

(LLM Stats (zeroeval))

Agents

AA-Briefcase1577.00 / 3000自報

Code

CursorBench v3.269.9%自報
DeepSWE 1.165.9%自報
APEX-SWE56.4%自報

General

GDPval-AA1753.00 / 3000自報
AA-Index61.0%自報

Knowledge

Harvey LAB (Vals)15.8%自報

Reasoning

FrontierCode61.3%自報
APEX-Agents57.5%自報
Terminal-Bench 3.026.0%自報
Terminal-Bench 4.020.3%

AA 評測指數

(Artificial Analysis)
Gpqa(NYU + Cohere + Anthropic (2023))
94.9
Terminalbench V2 1
88.4
Lcr(Artificial Analysis)
80.3
Coding Index(Artificial Analysis)
76.8
Scicode(UIUC + Argonne National Lab (2024))
56.5
Tau Banking
50.7
Intelligence Index(Artificial Analysis)
44.3
Hle(Center for AI Safety + Scale AI (2025))
42.9

LLM Stats 分類評分

(LLM Stats (zeroeval))
Legal
100
Finance
100
Productivity
100
Reasoning
100
Agents
100
General
100
Code
40
Tool Calling
20

定價

輸入價格$2 / 1M tokens
輸出價格$6 / 1M tokens
混合價格(3:1)$3 / 1M tokens
快取讀取價格$0.5 / 1M tokens

速度

Tokens/秒70.3
首Token延遲35.27s
首回答延遲35.27s

供應商價格排行

供應商價格排行

5 個供應商

最便宜: xAI最貴: Venice AI
供應商輸入輸出
1xAI最便宜
$0
$0.00001
2SpaceXAI主要
$2
$6
3DevPass (LLM Gateway)
$2
$6
4Neon
$2
$6
5Venice AI
$2.27
$6.8

比較該模型在不同 API 供應商之間的定價。

外部連結