Ling 3.0 Flash
InclusionAIUnknown
描述
The model prioritizes token efficiency and agentic inference at production scale, stretching what developers can achieve within limited token, latency, and serving-cost budgets.
发布日期
2026-08-04
参数规模
124.0B
上下文长度
262K
支持模态
text
能力雷达图
21
general
49
coding
86
reasoning
58
science
60
agents
0
multimodal
排行榜排名
基准测试分数 (LLM Stats)
(LLM Stats (zeroeval))Agents
DRACO
70.4%自报
Tau3 Banking
28.0%自报
Chat
Multi-IF
87.7%自报
Code
Artifacts Bench
77.0%自报
SkillsBench
44.8%自报
General
GDPval-AA
1107.00 / 3000自报
Instruction Following
IFBench
74.5%自报
Math
AIME 2026
93.2%自报
HMMT Feb 26
87.0%自报
IMO-AnswerBench
83.7%自报
Reasoning
LiveCodeBench v6
82.8%自报
SWE-bench Multilingual
72.4%自报
BrowseCompOpenAI (2025)
72.2%自报
MCP Atlas
65.5%自报
AA-LCR
65.1%自报
Terminal-Bench 2.1
57.0%自报
SWE-Bench ProPrinceton NLP (2024)
56.6%自报
Search
WideSearch
73.6%自报
Tool Calling
BFCL-V4
73.0%自报
AA 评测指数
(Artificial Analysis)Gpqa(NYU + Cohere + Anthropic (2023))85.5
Lcr(Artificial Analysis)73.0
Terminalbench V2 155.4
Coding Index(Artificial Analysis)50.6
Scicode(UIUC + Argonne National Lab (2024))42.0
Tau Banking27.2
Hle(Center for AI Safety + Scale AI (2025))23.7
Intelligence Index(Artificial Analysis)20.1
Terminalbench V4 00.0
LLM Stats 分类评分
(LLM Stats (zeroeval))Legal100
Finance100
Agents100
Reasoning75
General59
Chat90
Language90
Math90
Structured Output90
Communication90
Instruction Following80
Frontend Development80
Long Context70
Search70
Code60
Tool Calling60
定价
输入价格$0.075 / 1M tokens
输出价格$0.22 / 1M tokens
混合价格(3:1)$0.111 / 1M tokens
缓存读取价格$0.0042 / 1M tokens
速度
Tokens/秒331.5
首Token延迟1.82s
首回答延迟7.85s
供应商价格排行
供应商价格排行
2 个供应商
最便宜: DeepInfra最贵: InclusionAI
供应商输入输出
1DeepInfra最便宜
$0
$0
2InclusionAI主要
$0.075
$0.22
比较该模型在不同 API 供应商之间的定价。