T
traeai
Sign in

产品

NVIDIA Blackwell

别名:blackwell

NVIDIA推出的用于AI推理的GPU架构。

已跟踪 5 条高相关材料

TraeAI 观察

相关材料

已收录 5 条与 NVIDIA Blackwell 相关的内容,按评分排序。

CoreWeave Leads MLPerf 0.7 Endpoints Benchmark with DeepSeek-R1

CoreWeave Leads MLPerf 0.7 Endpoints Benchmark with DeepSeek-R1

CoreWeave1182 字 (约 5 分钟)
85

CoreWeave在MLPerf 0.7 Endpoints基准测试中,使用68块NVIDIA Blackwell GPU实现每秒44万输出token,验证了DeepSeek-R1模型的高效推理能力。

入选理由:CoreWeave在MLPerf 0.7测试中使用68块NVIDIA Blackwell GPU实现了441,740 token/s的吞吐量

FeaturedArticle#MLPerf#DeepSeek-R1#NVIDIA Blackwell#推理服务#基准测试英文
Optimize model training on Amazon SageMaker AI with NVIDIA Blackwell

Optimize model training on Amazon SageMaker AI with NVIDIA Blackwell

AWS Machine Learning Blog2771 字 (约 12 分钟)
85

NVIDIA Blackwell GPU与Amazon SageMaker AI结合,显著提升大规模AI模型训练效率,减少内存限制和通信开销。

入选理由:Blackwell的180GB内存和新型精度格式可减少内存压力,支持更大批次和更长序列。

FeaturedArticle#Amazon SageMaker#NVIDIA Blackwell#AI训练#机器学习英文
We just added significantly more NVIDIA Blackwell GPUs to better serve GLM-5.1 model on Ollama's clo...

Ollama Announces More NVIDIA Blackwell GPUs for GLM-5.1 Support

ollama(@ollama)202 字 (约 1 分钟)
75

Ollama adds more NVIDIA Blackwell GPUs to better support GLM-5.1 model in the cloud.

入选理由:Ollama 已扩展 NVIDIA Blackwell GPU 阵容以优化 GLM-5.1 模型性能。

FeaturedTweet#Ollama#GPU#GLM-5.1#NVIDIA#Cloud英文
DeepSeek-V4 Pro now available on Together AI

DeepSeek-V4 Pro Now Available on Together AI

Together AI Blog1895 字 (约 8 分钟)
75

Together AI launches DeepSeek-V4 Pro model with high-performance inference and multiple computing options.

入选理由:DeepSeek-V4 Pro 在 NVIDIA Blackwell 上实现 1.3 倍速度提升。

FeaturedArticle#AI#Model Deployment#Deep Learning中文
We’ve developed our own inference engine Runtime-Optimized Serving Engine (ROSE) to serve models ran...

We’ve developed our own inference engine ROSE

Perplexity(@perplexity_ai)302 字 (约 2 分钟)
65

Perplexity has launched its in-house inference engine ROSE, enabling efficient serving from embedding models to trillion-parameter LLMs, with CuTeDSL integration for faster GPU kernel customization.

入选理由:Perplexity 自主研发了推理引擎 ROSE,提升大模型服务效率。

FeaturedTweet#ROSE#CuTeDSL#GPU optimization#large model inference#Perplexity英文

跨材料问答 · NVIDIA Blackwell

回答基于:NVIDIA Blackwell 相关 5 条材料
    0 / 500

    AI may generate inaccurate information. Please verify important content.