Model Training & Fine-tuning

orchestra-research Bundle

Quantize LLMs to 8-bit or 4-bit for 50-75% memory reduction with minimal accuracy loss using bitsandbytes. Supports INT8, NF4, FP4 formats, QLoRA training, and 8-bit optimizers.

orchestra-research Bundle

Evaluates LLMs across 100+ benchmarks from 18+ harnesses (MMLU, HumanEval, GSM8K, safety, VLM) with multi-backend execution on local Docker, Slurm HPC, or cloud platforms.

orchestra-research Bundle

Generate high-quality sentence and text embeddings for semantic similarity, clustering, and retrieval using 5000+ pre-trained models. Supports multilingual and domain-specific embeddings for RAG and semantic search.

orchestra-research Bundle

Train and experiment with a minimal GPT implementation in ~300 lines of PyTorch, from character-level Shakespeare to GPT-2 scale.

orchestra-research Bundle

Manage and use Lambda Labs GPU cloud instances for ML training and inference with SSH access, persistent filesystems, and multi-node clusters.

orchestra-research Skill

Deploy Meta's LlamaGuard moderation model to filter LLM inputs and outputs across 6 safety categories using HuggingFace, vLLM, or FastAPI.

orchestra-research Bundle

Run LLM inference on CPU, Apple Silicon, and consumer GPUs without NVIDIA hardware. Use for edge deployment, M1/M2/M3 Macs, AMD/Intel GPUs, or when CUDA is unavailable. Supports GGUF quantization (1.5-8 bit) for reduced memory and 4-10× speedup vs PyTorch on CPU.

orchestra-research Bundle

GPU-accelerated data curation for LLM training, supporting text, image, video, and audio with fuzzy deduplication, quality filtering, semantic deduplication, PII redaction, and NSFW detection.

orchestra-research Bundle

Optimizes transformer attention with Flash Attention for 2-4x speedup and 10-20x memory reduction. Supports PyTorch native SDPA, flash-attn library, H100 FP8, and sliding window attention.

orchestra-research Bundle

Fine-tune and align language models using reinforcement learning with TRL, including SFT, DPO, PPO, GRPO, and reward model training.

orchestra-research Bundle

Expert guidance for implementing GRPO/RL fine-tuning with TRL for reasoning and task-specific model training.

orchestra-research Bundle

Optimizes LLM inference with NVIDIA TensorRT for maximum throughput and lowest latency on NVIDIA GPUs (A100/H100).

orchestra-research Bundle

Add distributed training support to any PyTorch script with minimal code changes using a unified API for DDP, DeepSpeed, FSDP, and mixed precision.

orchestra-research Bundle

Extend context windows of transformer models using RoPE, YaRN, ALiBi, and position interpolation techniques for processing long documents and implementing efficient positional encodings.

orchestra-research Bundle

Train Mixture of Experts (MoE) models using DeepSpeed or HuggingFace, covering architectures, routing, load balancing, and expert parallelism.

orchestra-research Bundle

Merge multiple fine-tuned models using mergekit to combine capabilities without retraining, covering SLERP, TIES-Merging, DARE, Task Arithmetic, linear merging, and production deployment strategies.

orchestra-research Skill

Train AI models to be harmless through self-critique and AI feedback using a set of constitutional principles, without requiring human labels for harmful outputs.

orchestra-research Bundle

Trains large language models (2B-462B parameters) using NVIDIA Megatron-Core with advanced parallelism strategies for maximum GPU efficiency.

orchestra-research Bundle

Adds PyTorch FSDP2 (fully_shard) to training scripts with correct init, sharding, mixed precision/offload config, and distributed checkpointing. Use when models exceed single-GPU memory or when you need DTensor-based sharding with DeviceMesh.

orchestra-research Bundle

Perform causal interventions on PyTorch models using pyvene's declarative framework for causal tracing, activation patching, and interchange intervention training.

orchestra-research Bundle

Run interpretability experiments on neural network internals using nnsight, with optional NDIF remote execution for massive models.

orchestra-research Bundle

Train and analyze Sparse Autoencoders (SAEs) using SAELens to decompose neural network activations into interpretable features for mechanistic interpretability research.

orchestra-research Bundle

Accelerate LLM inference using speculative decoding, Medusa multiple heads, and lookahead decoding techniques for 1.5-3.6× speedup without quality loss.

huggingface Skill

Plans and coordinates the deployment of a model to Amazon SageMaker AI, selecting the appropriate pathway (real-time, serverless, async, batch, or Bedrock CMI) based on model type, traffic, latency, and cost constraints.

antigravity Skill

Build production ML systems with PyTorch 2.x, TensorFlow, and modern ML frameworks, including model serving, feature engineering, A/B testing, and monitoring.

google-gemma Bundle

Selects the right Gemma model for a task, recommends deployment tooling (Gradio, Transformers.js, Vertex AI, MLX), and applies optimizations like MTP and QAT.

systemtce Bundle

Optimizes Dify workflows and plugins by restructuring graphs, reducing LLM token usage, tuning worker pools, and improving parallel processing.

majiayu000 Bundle

Covers AI, machine learning, LLMs, prompt engineering, and blockchain development with code examples and best practices for building AI applications and smart contracts.

majiayu000 Bundle

Guides machine learning development with experiment tracking, hyperparameter optimization, model registry, and MLOps pipeline integration.

majiayu000 Bundle

Annotates single-cell RNA-seq data by scoring marker genes, transferring labels with CellTypist, or reasoning over marker lists with an LLM.

majiayu000 Bundle

Quantize large language models to 4-bit precision using activation-aware weight quantization, reducing memory footprint and speeding up inference with minimal accuracy loss.

majiayu000 Bundle

Classifies HTML pages, forms, and fields using machine learning to detect page types, form types, and field types from HTML content or URLs.

majiayu000 Bundle

Quantize large language models to 8/4/3/2/1-bit precision without calibration data, using multiple optimized backends and integrations with HuggingFace Transformers, vLLM, and PEFT/LoRA.

majiayu000 Bundle

High-performance numerical computing with JAX, covering functional transformations, Flax NNX, and best practices for ML research.

majiayu000 Bundle

Scales Python ML workloads across clusters using Ray's distributed tasks, actors, data, and serving capabilities.

majiayu000 Bundle

Annotates single-cell RNA-seq data by scoring marker genes, transferring labels with CellTypist, or reasoning over cluster markers with an LLM.