Granite 3.3 8B (Non-reasoning)
IBM开源权重Apache 2.0 · 商用许可
描述
Granite-3.3-8B-Base is a decoder-only language model with a 128K token context window. It improves upon Granite-3.1-8B-Base by adding support for Fill-in-the-Middle (FIM) using specialized tokens, enabling the model to generate content conditioned on both prefix and suffix. This makes it well-suited for code completion tasks
发布日期
2025-04-16
参数规模
8.2B
上下文长度
—
支持模态
text
能力雷达图
16
general
12
coding
18
reasoning
20
science估算
16
agents
0
multimodal
Science 在缺少专门科学评测时使用推理能力代理估算。
排行榜排名
基准测试分数 (LLM Stats)
Code
HumanEval
89.7%自报
Creativity
AlpacaEval 2.0
62.7%自报
Arena Hard
57.6%自报
Finance
MMLU
63.9%自报
TruthfulQA
52.1%自报
General
TriviaQA
78.2%自报
IFEval
74.8%自报
ARC-C
50.8%自报
AGIEval
49.3%自报
NQ
36.5%自报
PopQA
26.2%自报
Language
Winogrande
74.4%自报
BIG-Bench Hard
69.1%自报
Math
AIME 2024
81.2%自报
MATH-500
69.0%自报
GSM8k
59.0%自报
DROP
36.1%自报
Reasoning
HumanEval+
86.1%自报
HellaSwag
80.1%自报
Safety
AttaQ
88.5%自报
AA 评测指数
Math Index6.7
Intelligence Index1.8
Math 5000.7
Mmlu Pro0.5
Gpqa0.3
Ifbench0.2
Livecodebench0.1
Tau20.1
Scicode0.1
Aime 250.1
Aime0.0
Lcr0.0
Hle0.0
Terminalbench Hard0.0
LLM Stats 分类评分
Safety90
Code90
Instruction Following70
Language70
Structured Output70
Legal60
Math60
Reasoning60
Finance60
General60
Healthcare60
Creativity60
Writing60
定价
输入价格$0.03 / 1M tokens
输出价格$0.25 / 1M tokens
混合价格(3:1)$0.085 / 1M tokens
速度
Tokens/秒369.4
首Token延迟21.86s
首回答延迟21.86s
供应商价格排行
供应商价格排行
1 个供应商
供应商输入输出
1IBM主要
$0.03
$0.25
比较该模型在不同 API 供应商之间的定价。