Heidelberg AI News

Latest Videos GitHub This week

Heidelberg AI News

All stories

1 story about dao-ailab/flash-attention

All Articles Videos

Any language English Deutsch

Friday 6 November

Zyphra ZYFA: FP8 FlashAttention for AMD MI355X, Up to 1.76x Faster
ArticleMarkTechPost
5 h

On GitHub

Repositories the news linked to in the last 14 days, newest first.

sgl-project/genai-bench SageMaker Inference now supports Multi-Instance GPU to maximize GPU utilization for model endpoints4 h ago
zyphra/zyfa
rocm/aiter
dao-ailab/flash-attention Zyphra ZYFA: FP8 FlashAttention for AMD MI355X, Up to 1.76x Faster5 h ago
awslabs/agentcore-samples Govern AI agent access with Okta Cross App Access and Amazon Bedrock AgentCore – Part 25 h ago
strands-agents/sdk-python Taxbit Assist: A grounded, role-aware AI agent on Amazon Bedrock AgentCore5 h ago
crgregersen/m10dis
autonomous-ai/personal-ai-computer [AINews] Inkling 2 Small: 17A-594B trained in 68 hours on 12T tokens, beating Muse Spark12 h ago
vercel-labs/agent-desktop HUGE Claude Sonnet 6 LEAK, Anthropic's SECRET Fable 6?! GPT-6.2 Astra TODAY, & More! AI NEWS13 h ago
thinking-machines-lab/tinker-cookbook Thinking Machines Releases Inkling 2 Small, a 594B MoE With 17B Active18 h ago
baidu/unlimited-ocr Kimi K3.1 weights OUT, open source computer use, SOTA image model, science model, Pika H3: AI NEWS1 day ago
hazyresearch/megakernels
mirage-project/mirage Together AI Releases Together Compiler: Agents Make Inference Up to 2.6x Faster2 days ago
bytedance/axion1_0
valeriodolci/frontier-weight-edit [AINews] It's Swapping Season2 days ago

Heidelberg AI News reads the AI labs and the creators who test their tools every hour and shows each story with its own opening lines and a link to the original.

RSS Sources Privacy Imprint info@32dots.de© 2026 32dots GmbH · v1.27.0