Results for “qc”

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.

Open-source embedding database for AI applications. Store embeddings and metadata, perform vector and full-text search, filter by metadata. Simple 4-function API. Scales from notebooks to production clusters. Use for semantic search, RAG applications, or document retrieval. Best for local development and open-source projects.

Multi-agent orchestration framework for autonomous AI collaboration. Use when building teams of specialized agents working together on complex tasks, when you need role-based agent collaboration with memory, or for production workflows requiring sequential/hierarchical execution. Built without LangChain dependencies for lean, fast execution.

Large Language and Vision Assistant. Enables visual instruction tuning and image-based conversations. Combines CLIP vision encoder with Vicuna/LLaMA language models. Supports multi-turn image chat, visual question answering, and instruction following. Use for vision-language chatbots or image understanding tasks. Best for conversational image analysis.

GPU-accelerated data curation for LLM training. Supports text/image/video/audio. Features fuzzy deduplication (16× faster), quality filtering (30+ heuristics), semantic deduplication, PII redaction, NSFW detection. Scales across GPUs with RAPIDS. Use for preparing high-quality training datasets, cleaning web data, or deduplicating large corpora.

Trains large language models (2B-462B parameters) using NVIDIA Megatron-Core with advanced parallelism strategies. Use when training models >1B parameters, need maximum GPU efficiency (47% MFU on H100), or require tensor/pipeline/sequence/context/expert parallelism. Production-ready framework used for Nemotron, LLaMA, DeepSeek.

Framework for building LLM-powered applications with agents, chains, and RAG. Supports multiple providers (OpenAI, Anthropic, Google), 500+ integrations, ReAct agents, tool calling, memory management, and vector store retrieval. Use for building chatbots, question-answering systems, autonomous agents, or RAG applications. Best for rapid prototyping and production deployments.

Runs 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.

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.

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.

Facebook's library for efficient similarity search and clustering of dense vectors. Supports billions of vectors, GPU acceleration, and various index types (Flat, IVF, HNSW). Use for fast k-NN search, large-scale vector retrieval, or when you need pure similarity search without metadata. Best for high-performance applications.

Post-training 4-bit quantization for LLMs with minimal accuracy loss. Use for deploying large models (70B, 405B) on consumer GPUs, when you need 4× memory reduction with <2% perplexity degradation, or for faster inference (3-4× speedup) vs FP16. Integrates with transformers and PEFT for QLoRA fine-tuning.

Quantizes LLMs to 8-bit or 4-bit for 50-75% memory reduction with minimal accuracy loss. Use when GPU memory is limited, need to fit larger models, or want faster inference. Supports INT8, NF4, FP4 formats, QLoRA training, and 8-bit optimizers. Works with HuggingFace Transformers.

Optimizes transformer attention with Flash Attention for 2-4x speedup and 10-20x memory reduction. Use when training/running transformers with long sequences (>512 tokens), encountering GPU memory issues with attention, or need faster inference. Supports PyTorch native SDPA, flash-attn library, H100 FP8, and sliding window attention.

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.

Parameter-efficient fine-tuning for LLMs using LoRA, QLoRA, and 25+ methods. Use when fine-tuning large models (7B-70B) with limited GPU memory, when you need to train <1% of parameters with minimal accuracy loss, or for multi-adapter serving. HuggingFace's official library integrated with transformers ecosystem.

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.

Activation-aware weight quantization for 4-bit LLM compression with 3x speedup and minimal accuracy loss. Use when deploying large models (7B-70B) on limited GPU memory, when you need faster inference than GPTQ with better accuracy preservation, or for instruction-tuned and multimodal models. MLSys 2024 Best Paper Award winner.

Generates publication-quality figures for ML papers from research context. Given a paper section or description, extracts system components and relationships to generate architecture diagrams via Gemini. Given experiment results or data, auto-selects chart type and generates data-driven figures via matplotlib/seaborn. Use when creating any figure for a conference paper.

Battle-tested PyTorch training recipes for all domains — LLMs, vision, diffusion, medical imaging, protein/drug discovery, spatial omics, genomics. Covers training loops, optimizer selection (AdamW, Muon), LR scheduling, mixed precision, debugging, and systematic experimentation. Use when training or fine-tuning neural networks, debugging loss spikes or OOM, choosing architectures, or optimizing GPU throughput.

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.

Fine-tune and serve Physical Intelligence OpenPI models (pi0, pi0-fast, pi0.5) using JAX or PyTorch backends for robot policy inference across ALOHA, DROID, and LIBERO environments. Use when adapting pi0 models to custom datasets, converting JAX checkpoints to PyTorch, running policy inference servers, or debugging norm stats and GPU memory issues.

Fine-tunes and evaluates OpenVLA-OFT and OpenVLA-OFT+ policies for robot action generation with continuous action heads, LoRA adaptation, and FiLM conditioning on LIBERO simulation and ALOHA real-world setups. Use when reproducing OpenVLA-OFT paper results, training custom VLA action heads (L1 or diffusion), deploying server-client inference for ALOHA, or debugging normalization, LoRA merge, and cross-GPU issues.

Orchestrates end-to-end autonomous AI research projects using a two-loop architecture. The inner loop runs rapid experiment iterations with clear optimization targets. The outer loop synthesizes results, identifies patterns, and steers research direction. Routes to domain-specific skills for execution, supports continuous agent operation via Claude Code /loop and OpenClaw heartbeat, and produces research presentations and papers. Use when starting a research project, running autonomous experiments, or managing a multi-hypothesis research effort.

Build, slice, concatenate, read, and write AnnData annotated data matrices (obs, var, X, layers, obsm, uns) — the scverse data STRUCTURE, not an analysis pipeline. Use when creating or wrangling .h5ad/zarr files, managing cell and gene annotations, concatenating batches, or handling layers/obsm/backed-mode; for the QC, normalization, clustering, UMAP, and differential-expression analysis pipeline prefer alterlab-scanpy instead, and for RNA velocity from spliced/unspliced layers prefer alterlab-scvelo instead. Part of the AlterLab Academic Skills suite.

Run RNA velocity analysis with scVelo on single-cell RNA-seq data — estimate cell-state transitions from spliced/unspliced mRNA dynamics, infer trajectory direction, compute latent time, and identify driver genes. Use when adding directionality to trajectories or studying differentiation dynamics from spliced/unspliced layers (velocyto/STARsolo output); for the general QC, clustering, UMAP, and differential-expression analysis pipeline prefer alterlab-scanpy instead, and for .h5ad data-structure I/O and layer wrangling prefer alterlab-anndata instead. Part of the AlterLab Academic Skills suite.

Run the standard single-cell RNA-seq analysis pipeline with Scanpy on AnnData — QC filtering, normalization, dimensionality reduction (PCA, UMAP, t-SNE), Leiden/Louvain clustering, marker/differential expression, PAGA trajectories, and plotting. Use when analyzing scRNA-seq data through clustering, cell-type annotation, DE, or pseudotime workflows; for building or reading the .h5ad data structure itself (layers, obs/var, concatenation, backed mode) prefer alterlab-anndata instead, and for RNA velocity from spliced/unspliced counts prefer alterlab-scvelo instead. Part of the AlterLab Academic Skills suite.

Apply the scGPT single-cell foundation model (Cui 2024) to annotate and embed cells — zero-shot and fine-tuned cell-type annotation, gene/cell embeddings, batch integration, and gene-regulatory / perturbation inference from AnnData. Use when annotating cell types with a pretrained foundation model, generating scGPT embeddings, integrating batches with a transformer, or running zero-shot single-cell inference on an h5ad. For probabilistic latent models (scVI/scANVI) prefer alterlab-scvi-tools; for the standard QC→cluster→UMAP→DE pipeline prefer alterlab-scanpy; for the AnnData data structure itself prefer alterlab-anndata; for protein language models prefer alterlab-esm. Part of the AlterLab Academic Skills suite.

Exploratory data analysis (EDA) on a scientific data file — auto-detects the format, runs structure/quality/statistics checks, and writes a markdown EDA report with downstream recommendations. Use when asked to "explore", "analyze", "summarize", "profile", or "QC" a data file, or to understand its structure/content/quality before deciding what analysis to run. Covers tabular (.csv .tsv .xlsx .parquet), arrays (.npy .npz .hdf5 .h5 .mat .fits), sequence/genomics (.fasta .fastq .sam .bam .vcf .bed .gff .gtf .h5ad), microscopy (.tif .nd2 .czi .lif .ims .dcm .nii), spectroscopy/MS (.mzML .mzXML .mgf .fid .jdx), chemistry (.pdb .cif .mol .sdf .xyz .gro), and proteomics/metabolomics (.pepXML .mzid .mzTab). For zero-shot forecasting of a series use alterlab-timesfm; to create/configure a chunked cloud array store use alterlab-zarr. Part of the AlterLab Academic Skills suite.