repoGitHubTrust 82 · PrimaryPublished yesterdayLive · yesterday
Deep-Spark/DeepSparkInference
DeepSparkInference has selected 216 inference models of both small and large sizes. The small models cover fields such as computer vision, natural language processing, and speech recognition; the LLMs involve various frameworks including vLLM, TGI and LMDeploy. This repository is the mirror of Gitee.
Lineage graph
Paper → model → repo connections mined from source citations (Tier-1 exact match).
Why these links exist
Every edge carries a method, confidence, and the source snippet that justified it — so bad links are debuggable.
- PossiblePossibly related (embedding) · 55%Understanding Large Language Models →
- PossiblePossibly related (embedding) · 52%When Structured Sparse Autoencoders Learn Consistent Concepts Across Modalities →
- PossiblePossibly related (embedding) · 51%Attend, Transform, or Silence: Operator-Level Visual Skipping for Efficient Multimodal LLM Inference →
- PossiblePossibly related (embedding) · 50%VendorBench-100: A Unified Cross-Paradigm Benchmark for Deepfake Image Detection →
- PossiblePossibly related (embedding) · 50%AnyGroundBench: A Specialized-Domain Benchmark for Video Grounding in Vision-Language Models →
- PossiblePossibly related (embedding) · 50%deepseek-ai/DeepSeek-R1 →
Implements
paperUnderstanding Large Language ModelspaperWhen Structured Sparse Autoencoders Learn Consistent Concepts Across ModalitiespaperAttend, Transform, or Silence: Operator-Level Visual Skipping for Efficient Multimodal LLM InferencepaperVendorBench-100: A Unified Cross-Paradigm Benchmark for Deepfake Image DetectionpaperAnyGroundBench: A Specialized-Domain Benchmark for Video Grounding in Vision-Language Models
Related to
Related across the graph
paperWhen Structured Sparse Autoencoders Learn Consistent Concepts Across ModalitiespaperVendorBench-100: A Unified Cross-Paradigm Benchmark for Deepfake Image DetectionpaperAttend, Transform, or Silence: Operator-Level Visual Skipping for Efficient Multimodal LLM Inferencemodeldeepseek-ai/DeepSeek-R1paperUnderstanding Large Language ModelspaperAnyGroundBench: A Specialized-Domain Benchmark for Video Grounding in Vision-Language Models