← All skills

emerging-techniques-knowledge-distillation

CommunityMIT

Compress large language models using knowledge distillation from teacher to student models. Use when deploying smaller models with retained performance, transferring GPT-4 capabilities to open-source models, or reducing inference costs. Covers temperature scaling, soft targets, reverse KLD, logit distillation, and MiniLLM training strategies.

The ZIP contains only this skill. It lives on GitHub Releases of the public orlandoneto/agent-skills repo — Ontoom is the showcase. A daily crawler refreshes the catalog.

How to install

  1. Download the ZIP and extract the skill folder.
  2. Cursor: .agents/skills/aitmpl-emerging-techniques-knowledge-distillation/ or .cursor/skills/aitmpl-emerging-techniques-knowledge-distillation/
  3. Claude Code: .claude/skills/aitmpl-emerging-techniques-knowledge-distillation/
  4. Other AIs: .agents/skills/aitmpl-emerging-techniques-knowledge-distillation/
cursorclaude-codecodexgeminiany

Other skills in this collection

Other collections

  • emerging-techniques-long-context

    Extend context windows of transformer models using RoPE, YaRN, ALiBi, and position interpolation techniques. Use when processing long documents (32k-128k+ tokens), extending pre-trained models beyond original context limits, or implementing efficient positional encodings. Covers rotary embeddings, attention biases, interpolation methods, and extrapolation strategies for LLMs.

  • emerging-techniques-model-merging

    Merge multiple fine-tuned models using mergekit to combine capabilities without retraining. Use when creating specialized models by blending domain-specific expertise (math + coding + chat), improving performance beyond single models, or experimenting rapidly with model variants. Covers SLERP, TIES-Merging, DARE, Task Arithmetic, linear merging, and production deployment strategies.

  • emerging-techniques-model-pruning

    Reduce LLM size and accelerate inference using pruning techniques like Wanda and SparseGPT. Use when compressing models without retraining, achieving 50% sparsity with minimal accuracy loss, or enabling faster inference on hardware accelerators. Covers unstructured pruning, structured pruning, N:M sparsity, magnitude pruning, and one-shot methods.

  • emerging-techniques-moe-training

    Train Mixture of Experts (MoE) models using DeepSpeed or HuggingFace. Use when training large-scale models with limited compute (5× cost reduction vs dense models), implementing sparse architectures like Mixtral 8x7B or DeepSeek-V3, or scaling model capacity without proportional compute increase. Covers MoE architectures, routing mechanisms, load balancing, expert parallelism, and inference optimization.

  • emerging-techniques-speculative-decoding

    Accelerate LLM inference using speculative decoding, Medusa multiple heads, and lookahead decoding techniques. Use when optimizing inference speed (1.5-3.6× speedup), reducing latency for real-time applications, or deploying models with limited compute. Covers draft models, tree-based attention, Jacobi iteration, parallel token generation, and production deployment strategies.

  • endpoint-validator

    Deterministic API endpoint validation with pass/fail reporting.