跳转到主要内容

ERNIE 4.5 300B A47B

BaiduProprietary

描述

ERNIE 4.5 is an enhanced version of ERNIE 4.0 with improved reasoning capabilities, better instruction following, and stronger performance across multiple domains including mathematics, coding, and creative writing. Features advanced knowledge integration, multilingual support, and optimized thinking capabilities. This 21B parameter model uses A3B quantization for efficient inference.

发布日期
2025-06-30
参数规模
21.0B
上下文长度
123K
支持模态
image, text

能力雷达图

29
general
43
coding
55
reasoning
48
science
51
agents
80
multimodal

排行榜排名

领域#排名分数来源
代码能力榜410
22.0
AA
通用能力榜422
30.0
AA
科学能力274
47.0
AA

基准测试分数 (LLM Stats)

(LLM Stats (zeroeval))

Biology

GPQANYU + Cohere + Anthropic (2023)74.0%自报

Factuality

SimpleQA1.8%自报

Finance

MMLU41.9%自报
MMLU-Pro16.0%自报

General

ARC-E60.7%自报
PIQA55.2%自报
MMLU-Redux43.2%自报
C-Eval40.7%自报
ARC-C40.6%自报
MBPP+40.2%自报
CMMLU39.8%自报
AGIEval28.5%自报

Language

Winogrande51.3%自报
CLUEWSC48.6%自报
BBH30.4%自报

Math

DROP28.6%自报
GSM8k25.2%自报
MATH12.4%自报

Reasoning

HellaSwagAI2 (2019)33.0%自报
HumanEval+25.0%自报

AA 评测指数

(Artificial Analysis)
Math Index(Artificial Analysis)
41.3
Intelligence Index(Artificial Analysis)
8.9
Math 500(OpenAI (2024), subset of Hendrycks et al. MATH (2021))
0.9
Gpqa(NYU + Cohere + Anthropic (2023))
0.8
Mmlu Pro(TIGER-Lab (Univ. of Waterloo, Toronto, CMU, 2024))
0.8
Aime(MAA (Mathematical Association of America))
0.5
Livecodebench(UC Berkeley + MIT + Cornell (2024))
0.5
Aime 25(MAA (Mathematical Association of America))
0.4
Ifbench(Google Research (2023))
0.4
Scicode(UIUC + Argonne National Lab (2024))
0.3
Terminalbench Hard(Stanford × Laude Institute (2026))
0.1
Lcr(Artificial Analysis)
0.0
Hle(Center for AI Safety + Scale AI (2025))
0.0
Tau2(Sierra + U Toronto + Vector Institute (2025))
0.0

LLM Stats 分类评分

(LLM Stats (zeroeval))
Biology
70
Chemistry
70
Physics
60
Reasoning
40
Language
40
General
40
Legal
30
Math
30
Finance
30
Healthcare
30
Factuality
0

定价

输入价格$0.28 / 1M tokens
输出价格$1.1 / 1M tokens
混合价格(3:1)$0.485 / 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

5 个供应商

最便宜: Baidu最贵: Kilo Gateway
供应商输入输出
1Baidu主要
$0.28
$1.1
2OpenRouter
$0.42
$1.25
3Jiekou.AI
$0.42
$1.25
4NovitaAI
$0.42
$1.25
5Kilo Gateway
$0.42
$1.25

比较该模型在不同 API 供应商之间的定价。

外部链接