Mercury 2
InceptionProprietary
描述
Mercury 2 is the fastest reasoning LLM, built on diffusion-based language model (dLLM) architecture. Instead of generating text token-by-token, it refines multiple text blocks simultaneously, achieving over 1,000 tokens per second on Nvidia Blackwell GPUs — 5x faster than leading speed-optimized LLMs. Supports tool usage and JSON output with 128K context window.
发布日期
2026-02-20
参数规模
—
上下文长度
128K
支持模态
text
能力雷达图
15
general
32
coding
77
reasoning
51
science
50
agents
0
multimodal
排行榜排名
基准测试分数 (LLM Stats)
(LLM Stats (zeroeval))Chat
Tau2 Airline
53.0%自报
Instruction Following
IFBench
71.0%自报
Math
AIME 2025
91.1%自报
Reasoning
GPQANYU + Cohere + Anthropic (2023)
74.0%自报
LiveCodeBench
67.0%自报
SciCode
38.0%自报
AA 评测指数
(Artificial Analysis)Gpqa(NYU + Cohere + Anthropic (2023))77.0
Tau2(Sierra + U Toronto + Vector Institute (2025))70.8
Ifbench(Google Research (2023))69.8
Lcr(Artificial Analysis)43.7
Scicode(UIUC + Argonne National Lab (2024))37.7
Coding Index(Artificial Analysis)31.1
Terminalbench V2 127.3
Terminalbench Hard(Stanford × Laude Institute (2026))26.5
Hle(Center for AI Safety + Scale AI (2025))17.1
Intelligence Index(Artificial Analysis)13.8
Tau Banking9.5
Terminalbench V4 00.0
LLM Stats 分类评分
(LLM Stats (zeroeval))Instruction Following70
General70
Math60
Physics60
Reasoning60
Biology60
Chemistry60
Chat50
Code50
Communication50
Tool Calling50
定价
输入价格$0.25 / 1M tokens
输出价格$0.75 / 1M tokens
混合价格(3:1)$0.375 / 1M tokens
缓存读取价格$0.025 / 1M tokens
速度
Tokens/秒0.0
首Token延迟0.00s
首回答延迟0.00s
供应商价格排行
供应商价格排行
6 个供应商
最便宜: Inception最贵: Venice AI
供应商输入输出
1Inception最便宜
$0
$0
2NanoGPT
$0.25
$0.75
3OpenRouter
$0.25
$0.75
4Kilo Gateway
$0.25
$0.75
5Vercel AI Gateway
$0.25
$0.75
6Venice AI
$0.3125
$0.9375
比较该模型在不同 API 供应商之间的定价。