Python · By Orchestra-Research

Runs lm-evaluation-harness to benchmark language models on academic suites such as MMLU, GSM8K and HumanEval, compare models and track training checkpoints.

Orchestra-Research/AI-Research-SKILLs

Guide to using Meta's Segment Anything Model for zero-shot image segmentation with point, box or mask prompts, or automatic mask generation.

Orchestra-Research/AI-Research-SKILLs

Shows how to store documents and embeddings in Chroma, query them by similarity with metadata filters, and persist them to disk for RAG and semantic search projects.

Orchestra-Research/AI-Research-SKILLs

Explains OpenAI's CLIP model for zero-shot image classification, image-text similarity, semantic image search and content moderation, with install steps and code patterns.

Orchestra-Research/AI-Research-SKILLs

Transcribes audio with OpenAI's Whisper: 99 languages, translation to English, language detection, six model sizes and word-level timestamps, from Python or the CLI.

Orchestra-Research/AI-Research-SKILLs

Sets up FAISS for fast nearest-neighbor search over large collections of dense vectors, choosing between Flat, IVF, HNSW and product quantization indexes.

Orchestra-Research/AI-Research-SKILLs

Shows how to load, train and use fast Hugging Face tokenizers, with BPE, WordPiece and Unigram models, padding, truncation and alignment tracking.

Orchestra-Research/AI-Research-SKILLs

Guide to LLaVA for image chat, visual question answering and captioning, with model sizes, CLI and Gradio usage and multi-turn conversation code.

Orchestra-Research/AI-Research-SKILLs

Shows how to organize PyTorch training with Lightning's LightningModule and Trainer, covering validation, DDP, callbacks and learning-rate scheduling.

Orchestra-Research/AI-Research-SKILLs

Teaches an agent to build LM pipelines, RAG systems and agents in DSPy using signatures, modules and optimizers instead of hand-tuned prompts.

Orchestra-Research/AI-Research-SKILLs

Uses the Outlines library to constrain model output to a JSON schema, Pydantic model, regex or fixed set of choices when running local models.

Orchestra-Research/AI-Research-SKILLs

Quick-reference help for fine-tuning language models with Axolotl, covering YAML configs, FSDP, context parallelism, compressed saves and dataset formats.

Orchestra-Research/AI-Research-SKILLs

Guides training and analyzing sparse autoencoders with SAELens to break neural network activations into interpretable features, including superposition and monosemanticity studies.

Orchestra-Research/AI-Research-SKILLs

Shows how to use Pinecone, a managed vector database, for production RAG, semantic search and recommendations: indexes, upserts, queries, filters and namespaces.

Orchestra-Research/AI-Research-SKILLs

Deploys LLMs with vLLM for high-throughput serving, covering the OpenAI-compatible server, offline batch inference, monitoring and a Docker rollout.

Orchestra-Research/AI-Research-SKILLs

Generates and edits images with Stable Diffusion through Hugging Face Diffusers, covering text-to-image, image-to-image, inpainting, SDXL and custom pipelines.

Orchestra-Research/AI-Research-SKILLs

Explains how to run Qdrant, a Rust vector database, for RAG and semantic search, covering collections, points, distance metrics and filtered or batched queries.

Orchestra-Research/AI-Research-SKILLs

Walks through aligning language models with SimPO, a reference-free preference optimization method, using accelerate configs for Mistral 7B, Llama 3 8B and math-focused models.

Orchestra-Research/AI-Research-SKILLs

Optimizes and serves LLMs on NVIDIA GPUs with TensorRT-LLM, covering quantization, in-flight batching, multi-GPU parallelism and the trtllm-serve command.

Orchestra-Research/AI-Research-SKILLs

Constrains language model output with regex, selections and grammars using the Guidance library, so JSON, XML, code or formatted fields come out valid.

Orchestra-Research/AI-Research-SKILLs

Adds distributed and mixed-precision training to a PyTorch script with a few Accelerate lines, then launches it on one GPU, many GPUs or DeepSpeed and FSDP setups.

Orchestra-Research/AI-Research-SKILLs

Guides mechanistic interpretability work with TransformerLens: loading models, caching activations, using HookPoints, activation patching and attention-pattern analysis.

Orchestra-Research/AI-Research-SKILLs

GGUF format and llama.cpp quantization for efficient CPU/GPU inference.

Orchestra-Research/AI-Research-SKILLs

Uses Ray Data to read, transform and write large datasets across a cluster for ML training and batch inference, with streaming execution and optional GPU steps.

Orchestra-Research/AI-Research-SKILLs

Guides causal experiments on PyTorch models with pyvene, such as causal tracing, activation patching and interchange intervention training, to test how a model works.

Orchestra-Research/AI-Research-SKILLs

Loads large language models in 8-bit or 4-bit with bitsandbytes so they fit smaller GPUs, and sets up QLoRA fine-tuning on a 4-bit base model.

Orchestra-Research/AI-Research-SKILLs

Scales PyTorch, TensorFlow and Hugging Face training from a single GPU to multi-node clusters with Ray Train, including Ray Tune sweeps and checkpoint recovery.

Orchestra-Research/AI-Research-SKILLs

Generates text embeddings locally with the sentence-transformers library for RAG, semantic search, clustering and similarity, with model picks for general, multilingual and legal text.

Orchestra-Research/AI-Research-SKILLs

Trains and uses SentencePiece tokenizers on raw text, with BPE or Unigram models, for multilingual and CJK projects that need a reproducible vocabulary.

Orchestra-Research/AI-Research-SKILLs

Covers serving LLMs with SGLang, whose RadixAttention reuses cached prefixes, and constraining output to JSON, regex or grammar for agent and tool-calling workloads.

Orchestra-Research/AI-Research-SKILLs

Explains three ways to speed up LLM inference: draft-model speculative decoding, Medusa heads and lookahead decoding with Jacobi iteration, and when each one fits.

Orchestra-Research/AI-Research-SKILLs

Guides reinforcement-learning research with torchforge, Meta's PyTorch-native library that keeps RL algorithms apart from infrastructure, including GRPO math-reasoning runs.

Orchestra-Research/AI-Research-SKILLs

Trains LLMs with reinforcement learning using verl, from ByteDance's Seed team, with GRPO, PPO and other algorithms and swappable training and rollout backends.

Orchestra-Research/AI-Research-SKILLs

Tracks ML experiments, versions models in the MLflow registry and covers deployment and reproducibility, with autologging for common frameworks.

Orchestra-Research/AI-Research-SKILLs

Walks through nanoGPT, Karpathy's compact GPT implementation: training on Shakespeare, reproducing GPT-2, fine-tuning GPT-2 checkpoints and training on your own text.

Orchestra-Research/AI-Research-SKILLs

Explains RWKV, a hybrid that trains in parallel like a GPT and runs inference like an RNN with constant memory per token, plus usage, fine-tuning and troubleshooting.

Orchestra-Research/AI-Research-SKILLs

Guidance for using A-Evolve to improve an AI agent automatically, evolving its prompts, skills and memory against a benchmark through solve, observe and evolve cycles.

Orchestra-Research/AI-Research-SKILLs

Shows how to log ML runs, configs, metrics and media with SwanLab and view them in cloud, local or self-hosted dashboards.

Orchestra-Research/AI-Research-SKILLs

Fine-tunes and evaluates OpenVLA-OFT and OFT+ robot policies with LoRA and continuous action heads on LIBERO simulation and ALOHA real-robot setups.

Orchestra-Research/AI-Research-SKILLs