Llama 3.3 Nemotron Super 49B v1 (Non-reasoning)
NVIDIALlama开源权重Llama 3.1 Community License
描述
Llama-3.3-Nemotron-Super-49B-v1 is a large language model (LLM) derived from Meta Llama-3.3-70B-Instruct. It's post-trained for reasoning, chat, RAG, and tool calling, offering a balance between accuracy and efficiency (optimized for single H100). It underwent multi-phase post-training including SFT and RL (RLOO, RPO).
发布日期
2025-03-18
参数规模
49.9B
上下文长度
—
支持模态
—
能力雷达图
80
general
0
coding
80
reasoning
60
science估算
70
agents
0
multimodal
缺少专门科学评测时,Science 由 LLM Stats 科学得分或推理能力估算。
排行榜排名
暂无排名数据
基准测试分数 (LLM Stats)
(LLM Stats (zeroeval))Chat
MT-Bench
0.92 / 100自报
General
Arena Hard
88.3%自报
BFCL v2
73.7%自报
Math
MATH-500
96.6%自报
AIME 2025
58.4%自报
Reasoning
MBPP
0.91 / 100自报
GPQANYU + Cohere + Anthropic (2023)
66.7%自报
AA 评测指数
(Artificial Analysis)暂无 AA 评测数据
LLM Stats 分类评分
(LLM Stats (zeroeval))Chat90
Roleplay90
Communication90
Creativity90
Writing90
Math80
Reasoning80
General80
Physics70
Biology70
Chemistry70
Tool Calling70
定价
暂无定价数据
速度
暂无速度数据
供应商价格排行
暂无提供商数据