VibeVoice

Google's pretrained foundation model for zero-shot time-series forecasting

TimesFM is a pretrained time-series foundation model from Google Research that performs zero-shot forecasting on diverse datasets without task-specific training. It handles univariate and multivariate time series across domains including finance, logistics, energy, and infrastructure monitoring with accuracy competitive against traditional statistical methods like ARIMA and Prophet.

First commercially viable 1-bit LLMs that are 14x smaller and 8x faster

PrismML Bonsai delivers the first commercially viable 1-bit large language models with 8B, 4B, and 1.7B parameter variants. The 8B model runs in just 1GB of RAM versus 16GB for standard FP16 models, achieving 44 tokens per second on iPhone. Backed by $16.25M from Khosla Ventures and released under Apache 2.0, Bonsai makes capable LLMs practical for edge devices and resource-constrained environments.

Production-grade reinforcement learning framework for LLM training

verl is an open-source reinforcement learning framework designed specifically for training and aligning large language models. Built for production use with support for distributed training across multiple GPUs and nodes, it implements RLHF, DPO, and other alignment algorithms that make LLMs follow instructions, avoid harmful outputs, and generate higher quality responses. Over 580 contributors and 20,000 GitHub stars signal strong adoption.

State-of-the-art open-source text-to-speech with emotion control

Chatterbox is an open-source text-to-speech model by Resemble AI that delivers state-of-the-art voice synthesis with fine-grained emotion and style control. The model supports zero-shot voice cloning from short audio samples, produces natural-sounding speech across multiple speaking styles, and runs locally without cloud dependencies. With over 24,000 GitHub stars, it has become the leading open-source alternative to commercial TTS services for developers building voice-enabled AI applications.

Kubernetes-native distributed LLM inference stack

llm-d is an open-source Kubernetes-native stack for distributed LLM inference with cache-aware routing and disaggregated serving. It separates prefill and decode stages across different GPU pools for optimal resource utilization, routes requests to nodes with warm KV caches, and integrates with vLLM as the serving engine. Apache-2.0 licensed with 2,900+ GitHub stars.

Open-source toolkit for curating datasets and evaluating visual AI models

FiftyOne is an open-source Python toolkit from Voxel51 for building high-quality datasets and better computer-vision and multimodal AI models. It pairs a browser-based visualization App with programmatic dataset curation, embeddings, similarity search, and model-evaluation workflows.

Private, self-hosted research notebooks with flexible AI models, source chat, and podcasts

Open Notebook is an MIT-licensed, self-hosted alternative to NotebookLM for collecting sources, chatting over research, generating reusable transformations, and producing multi-speaker podcasts. Its Docker stack keeps notebook data under the user's control while supporting 18-plus model providers, including local Ollama and LM Studio workflows.

Hugging Face's open-source inference server for embeddings, rerankers, and classifiers

Text Embeddings Inference is Hugging Face's Apache-2.0 server for high-throughput embedding, reranking, and sequence-classification models. TEI packages token-based dynamic batching, optimized Transformers kernels, Safetensors loading, OpenAI-compatible embedding endpoints, Prometheus metrics, and configurable OpenTelemetry tracing in deployable CPU and GPU images.

Open-source PII detection and anonymization for AI data flows

Presidio is an MIT-licensed privacy framework for identifying and anonymizing personally identifiable information in text, images, and structured data. It can act as a de-identification layer around LLM prompts, logs, RAG corpora, and customer-data workflows.

Lifelike AI voice generation, cloning, and voice agents

ElevenLabs is an AI voice platform for text-to-speech, voice cloning, and conversational AI agents, built on models like Multilingual v2 and the low-latency Flash v2.5 and Turbo v2.5. Developers call its API to generate lifelike narration, clone voices from short audio samples, dub content across 30+ languages, add sound effects, and deploy real-time voice agents for customer service, IVR, and interactive apps, with SDKs for Python, JavaScript, and more.

AI data runtime for multimodal datasets and vector search

Deep Lake is an open-source AI data runtime from Activeloop for storing, versioning, and querying multimodal data and embeddings. It fits teams building RAG, training, evaluation, or dataset-heavy agent workflows that need a bridge between vector search, structured metadata, and large image, text, audio, or video collections.