lukemcqueen/hermes-cortex - 7페이지

openclaw-migration

Import an OpenClaw setup (memories, skills) into Hermes.

원문 언어: 영어

huggingface-accelerate

Run PyTorch training across GPUs with minimal changes.

원문 언어: 영어

chroma

Embedding database for RAG and semantic search.

원문 언어: 영어

clip

Zero-shot image classification and image-text search.

원문 언어: 영어

faiss

Fast vector similarity search at billion scale.

원문 언어: 영어

optimizing-attention-flash

Speed up long-sequence transformer training and inference.

원문 언어: 영어

guidance

Constrain LLM output with grammars; guarantee valid JSON.

원문 언어: 영어

outlines

Outlines: structured JSON/regex/Pydantic LLM generation.

원문 언어: 영어

instructor

Structured LLM outputs validated with Pydantic.

원문 언어: 영어

lambda-labs-gpu-cloud

On-demand GPU cloud instances for ML training.

원문 언어: 영어

llava

Vision-language chat: VQA, captioning, image dialogue.

원문 언어: 영어

modal-serverless-gpu

Serverless GPU cloud for ML jobs and model APIs.

원문 언어: 영어

nemo-curator

Curate LLM training data: dedupe, filter, PII redaction.

원문 언어: 영어

peft-fine-tuning

Fine-tune large LLMs with LoRA on limited GPU memory.

원문 언어: 영어

pinecone

Managed vector DB for production RAG and search.

원문 언어: 영어

pytorch-fsdp

Fully sharded data-parallel training for large models.

원문 언어: 영어

pytorch-lightning

Clean training loops with built-in distributed support.

원문 언어: 영어

qdrant-vector-search

Vector search engine for production RAG systems.

원문 언어: 영어

sparse-autoencoder-training

Train sparse autoencoders to interpret model features.

원문 언어: 영어

simpo-training

Reference-free preference alignment, simpler than DPO.

원문 언어: 영어

slime-rl-training

RL post-training for LLMs with Megatron and SGLang.

원문 언어: 영어

stable-diffusion-image-generation

Text-to-image generation, inpainting, and img2img.

원문 언어: 영어

tensorrt-llm

High-throughput LLM inference on NVIDIA GPUs.

원문 언어: 영어

distributed-llm-pretraining-torchtitan

Pretrain LLMs at scale with PyTorch 4D parallelism.

원문 언어: 영어

fine-tuning-with-trl

TRL: SFT, DPO, GRPO, RLOO reward modeling for LLM RLHF.

원문 언어: 영어

canvas

Fetch Canvas LMS courses and assignments via API token.

원문 언어: 영어

here-now

Publish sites to {slug}.here.now and store files in Drives.

원문 언어: 영어

telephony

Provision Twilio numbers, SMS/MMS, and AI outbound calls.

원문 언어: 영어

darwinian-evolver

Evolve prompts/regex/SQL/code with Imbue's evolution loop.

원문 언어: 영어

duckduckgo-search

Free keyless web, news, and image search via ddgs.

원문 언어: 영어

gitnexus-explorer

Serve an interactive codebase knowledge graph web UI.

원문 언어: 영어

osint-investigation

Follow the money via public records and sanctions data.

원문 언어: 영어

searxng-search

Free keyless meta-search aggregating 70+ engines.

원문 언어: 영어

1password

Set up op CLI, sign in, and read or inject secrets.

원문 언어: 영어

change-test-loop

Small changes with real verification, bounded retries, self-healing. One change at a time. Never batch. Never skip verification. Triggers: 'fix tests', 'make tests pass', 'refactor', 'change-test loop', 'debug failing test'

원문 언어: 영어

session-manager

Session management skill — checkpoint/restore, context compression, progress tracking, and recovery for maintaining continuity across agent sessions.

원문 언어: 영어

cleanup-commit-regression-check

When scripts fail with NameError after a mass-edit commit.

원문 언어: 영어

arq-worker-startup-pitfalls

Use when an arq worker crash-loops or runs no jobs.

원문 언어: 영어

governance-compliance-reporting

Review agent commits for enforcement compliance.

원문 언어: 영어

health-external-verification

Verify your health endpoint is externally reachable by testing the URL end-to-end instead of assuming local-only based on process checks.

원문 언어: 영어