← all publishers

Orchestra Research

@orchestra-research source repo

Report this profile Claim this profile

• 114 skills
• 0 followers
• 10k repo stars
• 2 weeks ago last updated

1. ▌
   Compiler 2 · orchestra-research bundle

   Universal ARA Compiler. Converts ANY research input — PDF papers, GitHub repositories, experiment logs, code directories, raw notes, or combinations thereof — into a complete Agent-Native Research Artifact (ARA): a structured, machine-executable knowledge package with a cognitive layer (claims, concepts, methods), an artifact layer (code/configs/data as the work warrants), an exploration graph (research DAG), and grounded evidence. Works across any research field — not only model-training research. TRIGGERS: compile, create ARA, generate artifact, convert paper, build artifact, compile paper, ARA from PDF, ARA from repo, ARA from code, structure research, extract knowledge, extract figure data, digitize plot, read chart, figure to data
   10.4k repo stars
2. ▌
   Submit Ara 2 · orchestra-research bundle

   ARA Submitter. Takes a research directory, makes sure it is a valid Agent-Native Research Artifact (ARA) — compiling it with the `compiler` skill when it is not — guarantees it carries an interactive visualization (`research-visualizer`), then publishes it to the ARA Hub and hands back one link plus an update token. The token is stored in the artifact's own `.ara_env`, so the next submission of the same work REPLACES it in place instead of creating a second entry. Publishes through the user's GitHub account when they have one, and straight to the Hub when they do not — no account required either way. TRIGGERS: submit, submit ara, publish ara, upload ara, share ara, push ara to github, add to ara hub, submit-ara, publish artifact, make my ara public, submit to a conference, update my submission, resubmit
   10.4k repo stars
3. ▌
   Context Drop 2 · orchestra-research bundle

   Context Drop. Hands a file, a folder, or a set of notes to somebody else's agent as one URL. Uploads the path to the ARA Hub, then prints a share link plus a ready-to-paste prompt: the recipient's agent fetches the drop as a single Markdown document holding every text file, with binaries listed as URLs to pull on demand. Also the reader — given a drop link, it pulls the bundle and works from it. Replaces pushing a throwaway repo to GitHub or mailing a zip nobody's agent can open. TRIGGERS: context drop, drop, share this folder, share these files, send this directory, give this to my friend's agent, share context, make a link for this folder, upload this folder, share notes with an agent, read this drop, open a drop link, agenticresearch.sh/drop
   10.4k repo stars
4. ▌
   Rigor Reviewer 2 · orchestra-research bundle

   ARA Seal Level 2: Semantic Epistemic Review. Acts as an objective research reviewer for Agent-Native Research Artifacts. Assumes Level 1 structural validation has already passed. Evaluates six dimensions of epistemic quality through semantic reasoning over the ARA's content. Produces a scored review with per-dimension strengths/weaknesses/suggestions, severity-ranked findings, and an overall epistemic-quality tier (Exemplary to Unsound). TRIGGERS: level2, seal level 2, verify level 2, epistemic audit, review ara, audit claims
   10.4k repo stars
5. ▌
   Research Fuzzer 2 · orchestra-research bundle

   Treat an open-ended investigation the way a fuzzer treats a program. After every action, reflect on two things a fuzzer always knows and an agent never does: did anything NEW happen, and where have I NOT been yet. Keeps an append-only notebook of predictions and outcomes; reports what you explored, the leads you saw but never followed, unexplained results, and a going-in-circles alarm. Use for ANY investigation without a known map: research experiments, debugging, data analysis, literature or market research, evaluations. Fire it (1) when starting an investigation, (2) after every action or batch of actions that returned results, (3) before stating any conclusion. Skip it for trivial single-step tasks.
   10.4k repo stars
6. ▌
   Research Manager 2 · orchestra-research bundle

   End-of-turn research process recorder with progressive crystallization. Invoked at the END of EVERY turn, after the user's current request has been fully addressed and before yielding control back to the user. Reviews what happened in the turn, extracts research-significant events, and writes them into the ara/ artifact through a three-stage pipeline: Context Harvester → Event Router → Maturity Tracker. Trace events (decisions, experiments, dead ends, pivots) are recorded immediately as journey facts. Knowledge events (claims, heuristics, concepts, constraints) are staged first and crystallize into typed layers ONLY when closure signals appear — topic abandonment, verbal affirmation, empirical resolution, or artifact commitment. NEVER mid-turn. All entries carry provenance tags (user / ai-suggested / ai-executed / user-revised). Also supports optional, user-triggered taste comments — free-form evaluative reactions to a claim, heuristic, or trace node — independent of the crystallization pipeline.
   10.4k repo stars
7. ▌
   Research Foresight 2 · orchestra-research bundle

   ARA World Model — read-only reasoning engine over ONE Agent-Native Research Artifact (ARA), run LOCALLY with the coding agent itself as the LLM (no SDK, no API key). Given an ARA directory and a free-text query, it answers any question about the ARA — a forward "what if I change X", but equally why-did-this-work, what-should-I-try, is-this-sound, how-do-these-compare, or anything else — by retrieving precedent from the ARA's native files (references/RETRIEVE.md) and answering as the Predictor (references/PREDICT.md): a bold, grounded, falsifiable Answer shaped to what the question actually calls for. TRIGGERS: ask the world model, wm predict, predict with the world model, what if I change X, forecast the loss curve, will this help, why did this work, what should I try next, is this claim sound, compare these, retrieve precedent, what precedent surfaces.
   10.4k repo stars
8. ▌
   Research Visualizer 2 · orchestra-research bundle

   Research Visualizer. Renders an existing Agent-Native Research Artifact (ARA) into ONE self-contained, interactive HTML file showing the AI scientist's step-by-step research process: a clickable process map of the exploration tree (branches and dead ends included) on the left, and a per-step drill-down on the right — what the step did (its narrative written in plain language a person can follow), why (the linked claim), the real result (verbatim grounded numbers + inline figures + tables), and the code/artifact pointer. Read-only consumer of the artifact — it never changes how research is done. When the ARA carries them, it also surfaces (each optional, only when present) the related-work dependency graph, the problem framing, a concepts glossary with in-text term popovers, and the solution recipes — reached from header disclosures without leaving the process map. Accepts either an existing ARA or raw research input (a paper, repo, run logs, or notes); when the input is not yet an ARA it is compiled into one
   10.4k repo stars
9. ▌
   Compiler · orchestra-research bundle

   Universal ARA Compiler. Converts ANY research input — PDF papers, GitHub repositories, experiment logs, code directories, raw notes, or combinations thereof — into a complete Agent-Native Research Artifact (ARA): a structured, machine-executable knowledge package with a cognitive layer (claims, concepts, methods), an artifact layer (code/configs/data as the work warrants), an exploration graph (research DAG), and grounded evidence. Works across any research field — not only model-training research. TRIGGERS: compile, create ARA, generate artifact, convert paper, build artifact, compile paper, ARA from PDF, ARA from repo, ARA from code, structure research, extract knowledge, extract figure data, digitize plot, read chart, figure to data
   10.4k repo stars
10. ▌
    Submit Ara · orchestra-research bundle

    ARA Submitter. Takes a research directory, makes sure it is a valid Agent-Native Research Artifact (ARA) — compiling it with the `compiler` skill when it is not — guarantees it carries an interactive visualization (`research-visualizer`), then publishes it to the ARA Hub and hands back one link plus an update token. The token is stored in the artifact's own `.ara_env`, so the next submission of the same work REPLACES it in place instead of creating a second entry. Publishes through the user's GitHub account when they have one, and straight to the Hub when they do not — no account required either way. TRIGGERS: submit, submit ara, publish ara, upload ara, share ara, push ara to github, add to ara hub, submit-ara, publish artifact, make my ara public, submit to a conference, update my submission, resubmit
    10.4k repo stars
11. ▌
    Context Drop · orchestra-research bundle

    Context Drop. Hands a file, a folder, or a set of notes to somebody else's agent as one URL. Uploads the path to the ARA Hub, then prints a share link plus a ready-to-paste prompt: the recipient's agent fetches the drop as a single Markdown document holding every text file, with binaries listed as URLs to pull on demand. Also the reader — given a drop link, it pulls the bundle and works from it. Replaces pushing a throwaway repo to GitHub or mailing a zip nobody's agent can open. TRIGGERS: context drop, drop, share this folder, share these files, send this directory, give this to my friend's agent, share context, make a link for this folder, upload this folder, share notes with an agent, read this drop, open a drop link, agenticresearch.sh/drop
    10.4k repo stars
12. ▌
    Rigor Reviewer · orchestra-research bundle

    ARA Seal Level 2: Semantic Epistemic Review. Acts as an objective research reviewer for Agent-Native Research Artifacts. Assumes Level 1 structural validation has already passed. Evaluates six dimensions of epistemic quality through semantic reasoning over the ARA's content. Produces a scored review with per-dimension strengths/weaknesses/suggestions, severity-ranked findings, and an overall epistemic-quality tier (Exemplary to Unsound). TRIGGERS: level2, seal level 2, verify level 2, epistemic audit, review ara, audit claims
    10.4k repo stars
13. ▌
    Research Fuzzer · orchestra-research bundle

    Treat an open-ended investigation the way a fuzzer treats a program. After every action, reflect on two things a fuzzer always knows and an agent never does: did anything NEW happen, and where have I NOT been yet. Keeps an append-only notebook of predictions and outcomes; reports what you explored, the leads you saw but never followed, unexplained results, and a going-in-circles alarm. Use for ANY investigation without a known map: research experiments, debugging, data analysis, literature or market research, evaluations. Fire it (1) when starting an investigation, (2) after every action or batch of actions that returned results, (3) before stating any conclusion. Skip it for trivial single-step tasks.
    10.4k repo stars
14. ▌
    Research Manager · orchestra-research bundle

    End-of-turn research process recorder with progressive crystallization. Invoked at the END of EVERY turn, after the user's current request has been fully addressed and before yielding control back to the user. Reviews what happened in the turn, extracts research-significant events, and writes them into the ara/ artifact through a three-stage pipeline: Context Harvester → Event Router → Maturity Tracker. Trace events (decisions, experiments, dead ends, pivots) are recorded immediately as journey facts. Knowledge events (claims, heuristics, concepts, constraints) are staged first and crystallize into typed layers ONLY when closure signals appear — topic abandonment, verbal affirmation, empirical resolution, or artifact commitment. NEVER mid-turn. All entries carry provenance tags (user / ai-suggested / ai-executed / user-revised). Also supports optional, user-triggered taste comments — free-form evaluative reactions to a claim, heuristic, or trace node — independent of the crystallization pipeline.
    10.4k repo stars
15. ▌
    Research Foresight · orchestra-research bundle

    ARA World Model — read-only reasoning engine over ONE Agent-Native Research Artifact (ARA), run LOCALLY with the coding agent itself as the LLM (no SDK, no API key). Given an ARA directory and a free-text query, it answers any question about the ARA — a forward "what if I change X", but equally why-did-this-work, what-should-I-try, is-this-sound, how-do-these-compare, or anything else — by retrieving precedent from the ARA's native files (references/RETRIEVE.md) and answering as the Predictor (references/PREDICT.md): a bold, grounded, falsifiable Answer shaped to what the question actually calls for. TRIGGERS: ask the world model, wm predict, predict with the world model, what if I change X, forecast the loss curve, will this help, why did this work, what should I try next, is this claim sound, compare these, retrieve precedent, what precedent surfaces.
    10.4k repo stars
16. ▌
    Research Visualizer · orchestra-research bundle

    Research Visualizer. Renders an existing Agent-Native Research Artifact (ARA) into ONE self-contained, interactive HTML file showing the AI scientist's step-by-step research process: a clickable process map of the exploration tree (branches and dead ends included) on the left, and a per-step drill-down on the right — what the step did (its narrative written in plain language a person can follow), why (the linked claim), the real result (verbatim grounded numbers + inline figures + tables), and the code/artifact pointer. Read-only consumer of the artifact — it never changes how research is done. When the ARA carries them, it also surfaces (each optional, only when present) the related-work dependency graph, the problem framing, a concepts glossary with in-text term popovers, and the solution recipes — reached from header disclosures without leaving the process map. Accepts either an existing ARA or raw research input (a paper, repo, run logs, or notes); when the input is not yet an ARA it is compiled into one
    10.4k repo stars
17. ▌
    Brainstorming Research Ideas · orchestra-research

    Guides researchers through structured ideation frameworks to discover high-impact research directions.
    10.4k repo stars
18. ▌
    Ara Research Manager · orchestra-research bundle

    Records research provenance as a post-task epilogue, scanning conversation history to extract decisions, experiments, dead ends, claims, heuristics, and pivots, and writing them into the ara/ directory with provenance tags.
    10.4k repo stars
19. ▌
    Creative Thinking For Research · orchestra-research

    Applies cognitive science frameworks for creative thinking to CS and AI research ideation, using combinatorial creativity, analogical reasoning, constraint manipulation, and other empirically grounded strategies.
    10.4k repo stars
20. ▌
    Implementing Llms Litgpt · orchestra-research bundle

    Train, fine-tune, and deploy LLMs using LitGPT's clean implementations of 20+ architectures like Llama, Gemma, and Phi.
    10.4k repo stars
21. ▌
    Llama Factory · orchestra-research bundle

    Provides expert guidance for fine-tuning LLMs with LLaMA-Factory, covering WebUI no-code, 100+ models, 2/3/4/5/6/8-bit QLoRA, and multimodal support.
    10.4k repo stars
22. ▌
    Quantizing Models Bitsandbytes · 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.
    10.4k repo stars
23. ▌
    Nemo Evaluator Sdk · 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.
    10.4k repo stars
24. ▌
    Sentence Transformers · 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.
    10.4k repo stars
25. ▌
    Nanogpt · orchestra-research bundle

    Train and experiment with a minimal GPT implementation in ~300 lines of PyTorch, from character-level Shakespeare to GPT-2 scale.
    10.4k repo stars
26. ▌
    Sentencepiece · orchestra-research bundle

    Train and use SentencePiece tokenizers for multilingual NLP, supporting BPE and Unigram algorithms with raw Unicode text.
    10.4k repo stars
27. ▌
    Lambda Labs Gpu Cloud · 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.
    10.4k repo stars
28. ▌
    Llamaguard · orchestra-research

    Deploy Meta's LlamaGuard moderation model to filter LLM inputs and outputs across 6 safety categories using HuggingFace, vLLM, or FastAPI.
    10.4k repo stars
29. ▌
    Llama Cpp · 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.
    10.4k repo stars
30. ▌
    Guidance · orchestra-research bundle

    Control LLM output with regex and grammars to guarantee valid JSON, XML, or code generation, enforce structured formats, and build multi-step workflows using Microsoft Research's Guidance framework.
    10.4k repo stars
31. ▌
    Outlines · orchestra-research bundle

    Guarantee valid JSON, XML, or code structure during text generation using Pydantic models for type-safe outputs, supporting local models (Transformers, vLLM, llama.cpp) and maximizing inference speed with structured generation.
    10.4k repo stars
32. ▌
    Segment Anything Model · orchestra-research bundle

    Segment any object in images using points, boxes, or masks as prompts, or automatically generate all object masks with zero-shot transfer.
    10.4k repo stars
33. ▌
    Stable Diffusion Image Generation · orchestra-research bundle

    Generate images from text prompts, perform image-to-image translation, inpainting, and build custom diffusion pipelines using Stable Diffusion models via HuggingFace Diffusers.
    10.4k repo stars
34. ▌
    Nemo Curator · 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.
    10.4k repo stars
35. ▌
    Optimizing Attention Flash · 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.
    10.4k repo stars
36. ▌
    Distributed LLM Pretraining Torchtitan · orchestra-research bundle

    Pretrains large language models from scratch using PyTorch-native distributed training with 4D parallelism (FSDP2, TP, PP, CP) and Float8 support on H100 GPUs.
    10.4k repo stars
37. ▌
    Fine Tuning With Trl · orchestra-research bundle

    Fine-tune and align language models using reinforcement learning with TRL, including SFT, DPO, PPO, GRPO, and reward model training.
    10.4k repo stars
38. ▌
    Prompt Guard · orchestra-research

    Detect prompt injections and jailbreak attempts in LLM applications using Meta's 86M parameter classifier. Filter user inputs, third-party data, and RAG documents with low latency and multilingual support.
    10.4k repo stars
39. ▌
    Instructor · orchestra-research bundle

    Extract structured data from LLM responses with Pydantic validation, automatic retries, and streaming support across multiple providers.
    10.4k repo stars
40. ▌
    Grpo Rl Training · orchestra-research bundle

    Expert guidance for implementing GRPO/RL fine-tuning with TRL for reasoning and task-specific model training.
    10.4k repo stars
41. ▌
    Deepspeed · orchestra-research bundle

    Provides expert guidance for distributed training with DeepSpeed, covering ZeRO optimization stages, pipeline parallelism, FP16/BF16/FP8, 1-bit Adam, and sparse attention.
    10.4k repo stars
42. ▌
    Ray Train · orchestra-research bundle

    Scales machine learning training from single GPU to multi-node clusters with minimal code changes. Supports PyTorch, TensorFlow, and HuggingFace with built-in hyperparameter tuning, fault tolerance, and elastic scaling.
    10.4k repo stars
43. ▌
    Tensorrt LLM · orchestra-research bundle

    Optimizes LLM inference with NVIDIA TensorRT for maximum throughput and lowest latency on NVIDIA GPUs (A100/H100).
    10.4k repo stars
44. ▌
    Huggingface Accelerate · 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.
    10.4k repo stars
45. ▌
    Nemo Guardrails · orchestra-research

    Add programmable safety guardrails to LLM applications at runtime, including jailbreak detection, input/output validation, fact-checking, hallucination detection, PII filtering, and toxicity detection.
    10.4k repo stars
46. ▌
    Ml Training Recipes · orchestra-research bundle

    Provides battle-tested PyTorch training recipes for LLMs, vision, diffusion, and biomedical domains, covering training loops, optimizer selection, LR scheduling, mixed precision, and debugging.
    10.4k repo stars
47. ▌
    Evaluating Llms Harness · orchestra-research bundle

    Evaluates LLMs across 60+ academic benchmarks (MMLU, HumanEval, GSM8K, TruthfulQA, HellaSwag) using standardized prompts and metrics. Supports HuggingFace, vLLM, and API backends.
    10.4k repo stars
48. ▌
    Long Context · 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.
    10.4k repo stars
49. ▌
    Moe Training · orchestra-research bundle

    Train Mixture of Experts (MoE) models using DeepSpeed or HuggingFace, covering architectures, routing, load balancing, and expert parallelism.
    10.4k repo stars
50. ▌
    Model Merging · 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.
    10.4k repo stars
51. ▌
    Model Pruning · orchestra-research bundle

    Compress large language models by 40-60% with minimal accuracy loss using one-shot pruning techniques like Wanda and SparseGPT, enabling faster inference and deployment on constrained hardware.
    10.4k repo stars
52. ▌
    Ml Paper Writing · orchestra-research bundle

    Draft publication-ready ML/AI papers for top conferences (NeurIPS, ICML, ICLR, ACL, AAAI, COLM) by exploring research repositories, structuring arguments, verifying citations via APIs, and formatting submissions.
    10.4k repo stars
53. ▌
    Constitutional AI · orchestra-research

    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.
    10.4k repo stars
54. ▌
    Training Llms Megatron · orchestra-research bundle

    Trains large language models (2B-462B parameters) using NVIDIA Megatron-Core with advanced parallelism strategies for maximum GPU efficiency.
    10.4k repo stars
55. ▌
    Pytorch Fsdp2 · 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.
    10.4k repo stars
56. ▌
    Academic Plotting · orchestra-research bundle

    Generates publication-quality figures for ML papers, including architecture diagrams via Gemini and data-driven charts via matplotlib/seaborn.
    10.4k repo stars
57. ▌
    Huggingface Tokenizers · orchestra-research bundle

    Fast tokenization for NLP using Rust-based tokenizers supporting BPE, WordPiece, and Unigram algorithms, with training, alignment tracking, and padding/truncation.
    10.4k repo stars
58. ▌
    Pyvene Interventions · orchestra-research bundle

    Perform causal interventions on PyTorch models using pyvene's declarative framework for causal tracing, activation patching, and interchange intervention training.
    10.4k repo stars
59. ▌
    Nnsight Remote Interpretability · orchestra-research bundle

    Run interpretability experiments on neural network internals using nnsight, with optional NDIF remote execution for massive models.
    10.4k repo stars
60. ▌
    Sparse Autoencoder Training · orchestra-research bundle

    Train and analyze Sparse Autoencoders (SAEs) using SAELens to decompose neural network activations into interpretable features for mechanistic interpretability research.
    10.4k repo stars
61. ▌
    Evaluating Code Models · orchestra-research bundle

    Evaluates code generation models across HumanEval, MBPP, MultiPL-E, and 15+ benchmarks with pass@k metrics. Use when benchmarking code models, comparing coding abilities, testing multi-language support, or measuring code generation quality.
    10.4k repo stars
62. ▌
    Pytorch Lightning · orchestra-research bundle

    Organizes PyTorch code with a Trainer class, automatic distributed training (DDP/FSDP/DeepSpeed), callbacks, and minimal boilerplate. Scales from laptop to supercomputer with the same code.
    10.4k repo stars
63. ▌
    Systems Paper Writing · orchestra-research bundle

    Provides paragraph-level structural blueprints for writing systems papers targeting OSDI, SOSP, ASPLOS, NSDI, and EuroSys, including page allocation, writing patterns, and venue-specific checklists.
    10.4k repo stars
64. ▌
    Ara Compiler · orchestra-research bundle

    Compiles research inputs—PDFs, code, logs, notes—into structured Agent-Native Research Artifacts with cognitive and physical layers.
    10.4k repo stars
65. ▌
    Speculative Decoding · 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.
    10.4k repo stars
66. ▌
    Knowledge Distillation · orchestra-research bundle

    Compress large language models using knowledge distillation from teacher to student models, covering temperature scaling, soft targets, reverse KLD, logit distillation, and MiniLLM training strategies.
    10.4k repo stars
67. ▌
    Presenting Conference Talks · orchestra-research bundle

    Generates conference presentation slides (Beamer LaTeX PDF and editable PPTX) from a compiled paper with speaker notes and talk script.
    10.4k repo stars
68. ▌
    Transformer Lens Interpretability · orchestra-research bundle

    Inspect and manipulate transformer internals via HookPoints and activation caching for mechanistic interpretability research.
    10.4k repo stars
69. ▌
    Ara Rigor Reviewer · orchestra-research bundle

    Performs a semantic epistemic review of Agent-Native Research Artifacts, scoring six dimensions and producing a constructive report with a recommendation.
    10.4k repo stars
70. ▌
    Faiss · orchestra-research bundle

    Enables fast similarity search and clustering of dense vectors using FAISS, supporting billions of vectors, GPU acceleration, and various index types.
    10.4k repo stars
71. ▌
    Chroma · orchestra-research bundle

    Store and query embeddings with metadata filtering, vector search, and full-text search using an open-source database that scales from notebooks to production.
    10.4k repo stars
72. ▌
    Qdrant Vector Search · orchestra-research bundle

    Build production RAG and semantic search systems with a high-performance vector database written in Rust, supporting hybrid search, filtering, and horizontal scaling.
    10.4k repo stars
73. ▌
    Mlflow · orchestra-research bundle

    Track ML experiments, manage model registry with versioning, deploy models to production, and reproduce experiments with MLflow.
    10.4k repo stars
74. ▌
    Pinecone · orchestra-research bundle

    Provides code examples and best practices for using Pinecone, a managed vector database for production RAG, recommendation, and semantic search applications.
    10.4k repo stars
75. ▌
    Experiment Tracking Swanlab · orchestra-research bundle

    Track ML experiments with open-source run logging, local or self-hosted dashboards, and media visualization using SwanLab.
    10.4k repo stars
76. ▌
    Crewai Multi Agent · orchestra-research bundle

    Build teams of autonomous AI agents that collaborate to solve complex tasks using role-based delegation, memory, and sequential or hierarchical execution.
    10.4k repo stars
77. ▌
    Autogpt Agents · orchestra-research bundle

    Build, deploy, and manage continuous AI agents using a visual workflow editor or development toolkit.
    10.4k repo stars
78. ▌
    Evolving AI Agents · orchestra-research bundle

    Optimize AI agents through automated evolution cycles using LLM-driven mutation of prompts, skills, and memory against measurable benchmarks.
    10.4k repo stars
79. ▌
    Clip · orchestra-research bundle

    Enables zero-shot image classification, image-text matching, and cross-modal retrieval using OpenAI's CLIP model.
    10.4k repo stars
80. ▌
    Peft Fine Tuning · orchestra-research bundle

    Fine-tune large language models by training less than 1% of parameters using LoRA, QLoRA, and 25+ adapter methods, enabling efficient adaptation on consumer GPUs.
    10.4k repo stars
81. ▌
    Awq Quantization · orchestra-research bundle

    Quantize large language models to 4-bit using activation-aware weight quantization, achieving ~3x speedup with minimal accuracy loss for deployment on limited GPU memory.
    10.4k repo stars
82. ▌
    Hqq Quantization · orchestra-research bundle

    Quantize large language models to 8/4/3/2/1-bit precision without calibration data, using multiple optimized backends for deployment with vLLM or HuggingFace Transformers.
    10.4k repo stars
83. ▌
    Langchain · orchestra-research bundle

    Build LLM-powered applications with agents, chains, and RAG using a framework that supports multiple providers and 500+ integrations.
    10.4k repo stars
84. ▌
    Llava · orchestra-research bundle

    Enables visual instruction tuning and image-based conversations using open-source vision-language models. Supports multi-turn image chat, visual question answering, and image understanding tasks.
    10.4k repo stars
85. ▌
    Autoresearch · orchestra-research bundle

    Orchestrates end-to-end autonomous AI research projects using a two-loop architecture for rapid experimentation and synthesis, producing papers and presentations.
    10.4k repo stars
86. ▌
    Gguf Quantization · orchestra-research bundle

    Convert and quantize models to GGUF format for efficient CPU/GPU inference with llama.cpp, supporting 2-8 bit quantization and Apple Silicon acceleration.
    10.4k repo stars
87. ▌
    Gptq · orchestra-research bundle

    Quantize large language models to 4-bit with minimal accuracy loss using GPTQ, enabling deployment of 70B+ models on consumer GPUs with 4× memory reduction and 3-4× faster inference.
    10.4k repo stars
88. ▌
    Tensorboard · orchestra-research bundle

    Visualize training metrics, debug models with histograms, compare experiments, visualize model graphs, and profile performance using TensorBoard.
    10.4k repo stars
89. ▌
    Llamaindex · orchestra-research bundle

    Connects LLMs with user data for RAG applications, document Q&A, and knowledge retrieval using 300+ data connectors and vector indices.
    10.4k repo stars
90. ▌
    Blip 2 Vision Language · orchestra-research bundle

    Generate image captions, answer visual questions, and perform image-text retrieval using BLIP-2's Q-Former architecture with frozen vision encoders and LLMs.
    10.4k repo stars
91. ▌
    Fine Tuning Serving Openpi · orchestra-research bundle

    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.
    10.4k repo stars
92. ▌
    Verl Rl Training · orchestra-research bundle

    Train LLMs with reinforcement learning using verl (Volcano Engine RL), supporting RLHF, GRPO, PPO, and other algorithms for scalable post-training with flexible infrastructure backends.
    10.4k repo stars
93. ▌
    Whisper · orchestra-research bundle

    Transcribe and translate speech across 99 languages using OpenAI's Whisper model, with support for multiple model sizes, batch processing, and subtitle generation.
    10.4k repo stars
94. ▌
    Axolotl · orchestra-research bundle

    Provides expert guidance for fine-tuning LLMs with Axolotl, covering YAML configs, LoRA/QLoRA, DPO/KTO/ORPO/GRPO, and multimodal support.
    10.4k repo stars
95. ▌
    Unsloth · orchestra-research bundle

    Provides expert guidance for fast fine-tuning with Unsloth, including LoRA/QLoRA optimization, with 2-5x faster training and 50-80% less memory usage.
    10.4k repo stars
96. ▌
    Miles Rl Training · orchestra-research bundle

    Train large-scale MoE models with FP8/INT4 low-precision RL, speculative decoding, and train-inference alignment using the miles framework.
    10.4k repo stars
97. ▌
    Simpo Training · orchestra-research bundle

    Train language models with SimPO, a reference-free preference optimization method that outperforms DPO without needing a reference model.
    10.4k repo stars
98. ▌
    Slime Rl Training · orchestra-research bundle

    Post-train LLMs with reinforcement learning using the slime framework, which integrates Megatron-LM for training and SGLang for rollout generation.
    10.4k repo stars
99. ▌
    Modal Serverless Gpu · orchestra-research bundle

    Run ML workloads on Modal's serverless GPU cloud platform with auto-scaling, pay-per-second pricing, and Python-native infrastructure.
    10.4k repo stars
100. ▌
     Phoenix Observability · orchestra-research bundle

     Trace, evaluate, and monitor LLM applications with an open-source observability platform.
     10.4k repo stars