Packages
-
gpt-oss
Provides reference implementations for running…
unclear · aging · 88.1K/mo
-
hoptorch
Provides a safe wrapper around PyTorch's scan…
permissive · active · 87.9K/mo
-
mempalace
MemPalace stores conversation history and…
permissive · active · 87.8K/mo
-
keyphrase-vectorizers
Extracts keyphrases from text documents using…
permissive · dormant · 87.2K/mo
-
llama-index-llms-ibm
Integrates IBM watsonx.ai foundation models…
permissive · active · 87.2K/mo
-
openai-messages-token-helper
Estimates token usage for OpenAI Chat…
permissive · aging · 86.8K/mo
-
skope-rules
Skope-rules learns interpretable logical rules…
permissive · dormant · 86.6K/mo
-
autogluon.vision
Automated machine learning for image…
permissive · active · 86.2K/mo
-
aegra-api
Aegra API is a self-hosted Agent Protocol…
permissive · active · 86.0K/mo
-
trainer
Trainer is a PyTorch model training framework…
permissive · dormant · 85.9K/mo
-
llama-index-embeddings-ibm
Integrates IBM watsonx.ai embedding models with…
permissive · active · 85.6K/mo
-
amalgam-lang
Amalgam is a Python wrapper that lets you call…
agpl · active · 85.3K/mo
-
graphrag
GraphRAG extracts structured knowledge graphs…
permissive · active · 85.1K/mo
-
vllm-tpu
vllm-tpu is a high-throughput LLM inference and…
permissive · active · 85.1K/mo
-
torchsr
Provides pretrained neural network models for…
permissive · dormant · 85.1K/mo
-
axial-positional-embedding
Provides axial positional embeddings for…
permissive · aging · 84.7K/mo
-
autogluon.text
Automates machine learning for text data,…
permissive · active · 84.6K/mo
-
aip-agents-binary
A local execution agent library for building…
unclear · active · 84.6K/mo
-
torchcde
Provides differentiable GPU-capable solvers for…
permissive · aging · 84.4K/mo
-
snc4onnx
Merges multiple ONNX neural network models into…
permissive · active · 84.3K/mo
-
sog4onnx
Generates minimal ONNX operation graphs by…
permissive · dormant · 84.3K/mo
-
dtreeviz
dtreeviz renders decision trees from…
permissive · aging · 84.2K/mo
-
soa4onnx
Adds or modifies output nodes in ONNX model…
permissive · aging · 84.1K/mo
-
abstract-hugpy-dev
A self-hosted LLM console and OpenAI-compatible…
unclear · active · 83.9K/mo
-
snd4onnx
Removes specified nodes from ONNX model graphs…
permissive · aging · 83.4K/mo
-
lmcache
LMCache is a KV cache management layer that…
permissive · active · 83.2K/mo
-
sor4onnx
Renames input, output, and operator nodes in…
permissive · dormant · 83.1K/mo
-
verl
verl is a reinforcement learning training…
permissive · active · 83.0K/mo
-
memsearch
Memsearch builds a searchable semantic memory…
permissive · active · 83.0K/mo
-
onnx2json
Converts ONNX model files to JSON format,…
permissive · abandoned · 83.0K/mo
-
sam4onnx
Rewrites attributes and constants in ONNX model…
permissive · aging · 83.0K/mo
-
scc4onnx
Converts ONNX model input tensor dimensions and…
permissive · aging · 83.0K/mo
-
langchain-modal
Integrates Modal serverless compute sandboxes…
permissive · active · 82.9K/mo
-
json2onnx
Converts JSON files to ONNX model format,…
permissive · dormant · 82.9K/mo
-
scs4onnx
Compresses ONNX model files by deduplicating…
permissive · abandoned · 82.9K/mo
-
awslabs.bedrock-kb-retrieval-mcp-server
An MCP server that connects AI tools to Amazon…
permissive · active · 82.9K/mo
-
promptflow
Prompt flow is a development suite for…
permissive · active · 82.8K/mo
-
ssc4onnx
Analyzes and displays the structure of ONNX…
permissive · dormant · 82.8K/mo
-
adapters
Adapters adds parameter-efficient fine-tuning…
permissive · active · 82.8K/mo
-
sio4onnx
Modifies the input and output shapes of ONNX…
permissive · aging · 82.8K/mo
-
sbi4onnx
Modifies ONNX model files to initialize or…
permissive · dormant · 82.7K/mo
-
ssi4onnx
Infers and fills missing tensor shape…
permissive · dormant · 82.7K/mo
-
sna4onnx
Adds new operations (nodes) to ONNX model…
permissive · abandoned · 82.7K/mo
-
spo4onnx
Applies specialized optimization techniques to…
permissive · dormant · 82.7K/mo
-
soc4onnx
Changes the opset version of an ONNX model…
permissive · abandoned · 82.7K/mo
-
sed4onnx
Encodes and decodes Base64-formatted constant…
permissive · abandoned · 82.7K/mo
-
sod4onnx
sod4onnx removes specified output operations…
permissive · abandoned · 82.6K/mo
-
sde4onnx
Removes embedded doc_strings from ONNX model…
permissive · abandoned · 82.6K/mo
-
CoLT5-attention
Implements conditionally routed efficient…
permissive · active · 82.5K/mo
-
svs4onnx
Swaps connections between output and input…
permissive · abandoned · 82.5K/mo
-
deepagents-cli
Command-line tool for scaffolding, configuring,…
permissive · active · 82.3K/mo
-
flashoptim
FlashOptim provides drop-in replacements for…
permissive · active · 82.2K/mo
-
llama-index-llms-gemini
Integrates Google's Gemini LLM with LlamaIndex,…
permissive · aging · 82.0K/mo
-
panns-inference
Provides pretrained neural network models for…
permissive · dormant · 81.8K/mo
-
sae-lens
SAE Lens trains and analyzes sparse…
permissive · active · 81.7K/mo
-
llama-index-embeddings-vertex
Integrates Google Vertex AI embedding models…
permissive · active · 81.6K/mo
-
livekit-plugins-inworld
Integrates Inworld's text-to-speech and…
permissive · active · 81.6K/mo
-
qrdet
Detects and segments QR codes in images using a…
permissive · aging · 81.6K/mo
-
llama-index-llms-litellm
Integrates LiteLLM with LlamaIndex to provide…
permissive · active · 81.3K/mo
-
pandasai-litellm
Integrates LiteLLM with PandasAI to enable…
permissive · aging · 81.1K/mo
-
transformer-engine-cu13
Accelerates Transformer model training and…
unclear · active · 81.0K/mo
-
liac-arff
Reads and writes ARFF (Attribute-Relation File…
permissive · abandoned · 80.9K/mo
-
hindsight-api-slim
Hindsight-api-slim provides a persistent memory…
permissive · active · 80.6K/mo
-
mlserver-mlflow
Provides a MLServer runtime that loads and…
permissive · aging · 80.3K/mo
-
llm-openai-plugin
A plugin that extends LLM to access OpenAI…
permissive · aging · 80.2K/mo
-
farm-haystack
Haystack is an end-to-end NLP framework for…
permissive · active · 80.2K/mo
-
tpu-inference
tpu-inference is a hardware plugin for vLLM…
permissive · active · 79.2K/mo
-
torchgeo
TorchGeo provides PyTorch datasets, samplers,…
permissive · active · 79.2K/mo
-
llama-index-tools-mcp
Connects LlamaIndex agents to MCP (Model…
permissive · active · 78.9K/mo
-
fa3-fwd
Provides a lightweight Flash-Attention-3…
permissive · active · 78.9K/mo
-
agentrust-trace
Creates and verifies cryptographically signed…
permissive · active · 78.7K/mo
-
pytorch-pretrained-bert
Provides PyTorch implementations of BERT, GPT,…
permissive · active · 78.6K/mo
-
perceptron
Perceptron is a Python SDK for building…
unclear · active · 78.5K/mo
-
dstack
dstack is a unified control plane for…
copyleft · active · 78.0K/mo
-
azure-ai-language-questionanswering
Python client library for Azure's Question…
permissive · active · 77.7K/mo
-
stop-sequencer
Adds stop-sequence support to Huggingface…
unclear · abandoned · 77.6K/mo
-
tavily-cli
Tavily CLI provides command-line and…
permissive · active · 77.2K/mo
-
clarifai
Official Python client for Clarifai's AI…
permissive · active · 76.7K/mo
-
e3nn-jax
Implements equivariant convolutional neural…
permissive · active · 76.6K/mo
-
cpex
CPEX is a plugin framework that lets you…
permissive · active · 76.5K/mo
-
megatron-fsdp
Megatron-FSDP implements Fully Sharded Data…
permissive · active · 76.4K/mo
-
azureml-contrib-services
Provides utilities for handling raw HTTP…
unclear · active · 76.4K/mo
-
pocket-tts
Pocket TTS generates speech from text on CPU…
unclear · active · 76.3K/mo
-
zuko
Zuko implements normalizing flows in PyTorch as…
permissive · active · 76.3K/mo
-
lovely-tensors
Monkey-patches PyTorch tensors to display…
permissive · active · 75.9K/mo
-
google-metrax
Metrax provides standard evaluation metrics for…
permissive · active · 75.6K/mo
-
vllm-sr
Routes LLM requests to appropriate models in a…
permissive · active · 75.6K/mo
-
microsoft-fabric-rti-mcp
An MCP server that bridges AI agents to…
permissive · active · 75.2K/mo
-
livekit-plugins-anam
Anam plugin for the LiveKit Agents framework…
permissive · active · 75.0K/mo
-
qonnx
QONNX provides Python utilities to work with…
permissive · active · 74.0K/mo
-
azure-cognitiveservices-knowledge-qnamaker
Provides a Python client library for…
permissive · active · 73.7K/mo
-
llama-index-vector-stores-lancedb
Integrates LanceDB as a vector store backend…
permissive · active · 73.5K/mo
-
mrmr-selection
Implements the mRMR (minimum Redundancy -…
unclear · dormant · 73.5K/mo