3dgs Training Debugger

Diagnose and fix 3DGS training-time failures: NaN losses, OOM crashes, divergent optimization, floater artifacts, densification failures, hyperparameter sensitivity. Covers runtime debugging for vanilla 3DGS and 50+ novel methods (deformable, MoE, physics-based, feed-forward). Detects 60 runtime failure patterns. Use when: 3DGS training crashes or produces poor results, loss is NaN/Inf, VRAM exhaustion, Gaussians explode or vanish, densification not working, convergence stalls, 训练调试/显存溢出/训练发散/浮点伪影.

188 stars 13 forksSoftware Development Plugin marketplace

Install

Claude Code plugin
/plugin marketplace add jaccen/Awesome-Gaussian-Skills

The repo is a Claude Code plugin marketplace. Run /plugin afterwards to pick the plugin that includes this skill.

skills CLI (Claude Code, Cursor, Codex and more)
npx skills add jaccen/Awesome-Gaussian-Skills --skill 3dgs-training-debugger
Manual install
git clone --depth 1 https://github.com/jaccen/Awesome-Gaussian-Skills.git /tmp/Awesome-Gaussian-Skills
mkdir -p ~/.claude/skills
cp -r /tmp/Awesome-Gaussian-Skills/skills/3dgs-training-debugger ~/.claude/skills/3dgs-training-debugger

Personal skills live in ~/.claude/skills. Use .claude/skills in a project folder to share a skill with your team.

From the SKILL.md

# 3DGS Training Debugger

You are a senior 3DGS engineer who has trained hundreds of Gaussian Splatting models across vanilla 3DGS, deformable GS, feed-forward GS, SLAM-GS, and physics-based GS pipelines. Diagnose and fix training-time failures systematically.

## Capabilities

- Diagnose training crashes (OOM, NaN/Inf loss, CUDA errors) with root-cause analysis
- Identify convergence failures (stalls, divergence, premature plateau)
- Debug densification failures (over/under-triggering, positional gradient issues)
- Diagnose visual artifacts from training logs (floaters, blur, ghosting, holes)
- Recommend hyperparameter adjustments with expected impact
- Guide distributed/multi-GPU training setup and debugging
- Troubleshoot checkpoint save/resume issues
- Address novel method stability (deformable GS, MoE-GS, physics-based GS, feed-forward GS)
- Detect 60 runtime failure patterns (see references/runtime-bug-patterns.md)

## Relationship to Other Skills

This skill covers the **runtime training phase** — what happens AFTER code is written and BEFORE evaluation. It complements:

- **3dgs-code-reviewer**: Static code analysis (pre-training). Use code-reviewer first to catch implementation bugs, then use this skill to debug runtime issues.
- **3dgs-experiment-planner**: Experiment design (pre-training). Design experiments, then use this skill when training fails.
Read the full skill on GitHub

Details

Skill name
3dgs-training-debugger
Path
skills/3dgs-training-debugger/SKILL.md
License
Apache-2.0
Repo updated
18 November 2026

FAQ

What does the 3dgs Training Debugger skill do?

Diagnose and fix 3DGS training-time failures: NaN losses, OOM crashes, divergent optimization, floater artifacts, densification failures, hyperparameter sensitivity. Covers runtime debugging for vanilla 3DGS and 50+ novel methods (deformable, MoE, physics-based, feed-forward). Detects 60 runtime failure patterns. Use when: 3DGS training crashes or produces poor results, loss is NaN/Inf, VRAM exhaustion, Gaussians explode or vanish, densification not working, convergence stalls, 训练调试/显存溢出/训练发散/浮点伪影.

How do I install the 3dgs Training Debugger skill?

Claude Code plugin: /plugin marketplace add jaccen/Awesome-Gaussian-Skills. Skills go in ~/.claude/skills (personal) or .claude/skills (project), and Claude loads them automatically when a task matches the description.

Is 3dgs Training Debugger popular?

It lives in jaccen/Awesome-Gaussian-Skills, which has 188 GitHub stars and 13 forks.

Related skills

3dgs Articulated Reasoner

188

jaccen/Awesome-Gaussian-Skills

3DGS Articulated Object Reasoning & Digital Twin Agent. Reason about articulated objects in 3DGS scenes: extract part structure, infer kinematic constraints, generate interactive digital twins. Use when: articulated object, digital twin, part structure, kinematic chain, URDF, interactive 3DGS, part-aware rendering, articulated manipulation, joint estimation, part segmentation 3DGS, ArtiSplat, ArtiTwinSplat, articulated reconstruction.

AI, LLMs & Agents13

3dgs Code Reviewer

188

jaccen/Awesome-Gaussian-Skills

Review and author 3DGS implementation code for correctness, performance, and best practices. Covers CUDA kernels, rendering pipeline, training loop, loss functions, gsplat/Nerfstudio migration, large-scale engineering, AI/LLM fusion, differentiable physics, web/mobile deployment, and general graphics math. Detects 164 known bug patterns including compression, forensics, SLAM, feed-forward, method-specific failures, environment/compilation, code authoring, and graphics math errors. Use when: reviewing or writing 3DGS/Gaussian Splatting CUDA code, debugging rendering artifacts, optimizing 3DGS t

Software Development13

3dgs Compression Deploy

188

jaccen/Awesome-Gaussian-Skills

3DGS compression-to-deployment pipeline: quantization (scalar/VQ/mixed-precision), pruning (coreset/adaptive/variational/merge/Bayesian), progressive streaming & LoD, Web/WebGPU/mobile deployment, hardware acceleration (Tensor Core/GEMM/FPGA/ASIC), training-free semantic compression. Covers 53+ methods across 6 compression categories. Use when: compressing 3DGS models, deploying 3DGS to web/mobile/edge, selecting quantization bit-width, designing streaming pipelines, 3DGS压缩/量化/剪枝/部署/流式传输/移动端/硬件加速.

DevOps & Cloud13

Patent Software Ip

188

jaccen/Awesome-Gaussian-Skills

Generate CN patent docs (claims, specification, abstract) and software copyright materials from AI/big-data project code or docs. Covers 7 AI domains + big data, 11 claim templates, auto domain detection, desensitization, prior-art search, and self-check. Use when: drafting CN patents from AI code/docs, generating software copyright materials, prior-art search for AI inventions, 专利撰写/软件著作权/AI知识产权/脱敏处理.

Software Development13

Review Architecture

155

serpro69/claude-toolbox

Review a written architecture artifact — an ADR (docs/adr/), a broader architecture doc, the architecture section of a design doc, or a domain-reference kit (glossary + traps pages, reviewed as one composite artifact) — against the system it claims to describe. Verifies the EXISTENCE and TOPOLOGY of declared mechanisms (structural boundaries, data ownership, NFR mechanisms, failure isolation, state consistency, evolution/versioning, domain binding) plus decision soundness, reversibility, and provenance (self-certification of reverse-engineered claims). Use after an ADR, architecture doc, domai

Software DevelopmentIn 1 list30

S4h Writing Character Development

242

human-avatar/skills-for-humanity

Engineers psychologically compelling characters by mapping want vs. need, wound, defence mechanism, and defining contradiction. Use when a character feels flat, underdeveloped, or unconvincing. Triggers: 'my character feels flat', 'character development', 'make this character feel real', 'this character isn't working', 'character work', 'how do I develop this character'.

Software DevelopmentIn 1 list25

Browse the most popular Claude skills, all software development skills, or write your own with the skills generator.