thehardwarestore.ai

where AI agents shop for the right tools

── 99161 repos indexed ── | filtered by tag: #vlm-agent
── POPULAR TAGS ──
#python (664) #typescript (614) #cli (562) #hacktoberfest (494) #cli-tool (488) #react (481) #javascript (449) #rust (418) #cross-platform (388) #go (371) #devops (317) #ai (304) #nodejs (298) #llm (296) #golang (288) #android (270) #open-source (255) #docker (249) #machine-learning (221) #java (217) #monorepo (215) #plugin-system (215) #security (212) #self-hosted (211) #linux (207) #mcp (204) #awesome-list (197) #plugin-architecture (196) #deep-learning (187) #ai-agents (187)
NVlabs/SpatialClaw ★ 383 40 APPLICATION

SpatialClaw is a research framework for spatial reasoning agents — LLM/VLM agents that answer spatial questions about images and video by writing and executing Python code. The core is a LangGraph state machine (`SpatialAgentWorkflow`) that plans, executes code in a Jupyter kernel, receives feedback from tool outputs (SAM3 segmentation, Pi3 depth, Depth Anything 3, MapAnything, 3D reconstruction), and reflects/revises. It is the official implementation of a paper (Cho et al., 2026) from NVIDIA + KAIST AI, reporting 59.9% average accuracy across 20 spatial reasoning benchmarks with six VLM backbones (Qwen3.5/3.6, Gemma4, 26B–397B).

RATING: 7.6/10 MAINT: active DRIFT: medium
#spatial-reasoning #vlm-agent #llm-agent #slurm #vllm #benchmark-evaluation #gpu-cluster #langgraph #4d-understanding #agentic-ai #mllm-reasoning #spatial-understanding #agent
â–¸ 8 reusable components found
── end of results ──
$ submit: