Language Models

A demo repo for running extreme-quantized language models locally without needing a research cluster.

It breaks complex tasks across a team of specialized LLM agents that refine their own skills as they work.

It split off from `verl` to give diffusion, video, and omni-modality models an RL post-training framework that doesn't treat them like chatbots.

NeMo AutoModel automates the busywork of wiring HuggingFace LLMs and VLMs into PyTorch-native distributed training so you can fine-tune or pretrain without hand-rolling parallelism code.

Xime is a deliberately minimal, Rime-based Android input method that serves as its author's personal testbed for on-device AI experiments in predictive text and speech recognition.

It turns your local machine into an OpenAI-compatible inference endpoint so agents and IDEs can run on offline models without reconfiguration.

Uses multimodal LLMs to transcribe PDFs into Markdown, preserving complex layouts that traditional extractors mangle.

xLLM is a C++ inference framework specifically optimized for Chinese AI accelerators, and it already powers JD.com’s core retail production workloads.

Curated technical deep-dives covering everything from NVLink signal integrity to Kubernetes GPU scheduling and Huawei NPU porting.

AutoFlow turns documentation into a conversational knowledge graph, then lets you embed the chat window anywhere.

Bytez wraps 175,000+ AI models behind a single endpoint so you don't have to host them yourself.

OpenMythos is an independent attempt to reconstruct Anthropic’s rumored Claude Mythos architecture as a trainable Recurrent-Depth Transformer with switchable attention and sparse MoE layers.

HRM-Text claims to cut pretraining costs by 130–600× compute and 150–900× data, shipping a full 1B-parameter framework with FSDP2, FlashAttention 3, and a hierarchical recurrent architecture.

MiniMind-O packs listen-see-speak intelligence into a 0.1B-parameter model you can retrain from the first line of code on a single desktop GPU.

It turns the scattered firehose of agentic AI tools into an 8-stage curriculum with homework and realistic time budgets.

Apfel surfaces Apple’s built-in FoundationModels as a pipe-friendly UNIX tool and OpenAI-compatible local server, no API keys required.

MTPLX squeezes extra tokens per second out of Apple Silicon by using the multi-token prediction heads that ship with modern models like Qwen 3.6, instead of leaving them idle like most runtimes.

A modular speech-to-speech pipeline that exposes an OpenAI Realtime-compatible WebSocket API so you can run voice agents on local or open-source models instead of proprietary cloud services.

This Go CLI turns a single sentence into a full novel by making Architect, Writer, and Editor LLM agents plan, draft, and review inside a long-loop state machine—no human hand-holding required.

It corrals the latest subquadratic sequence-model research into hardware-efficient, training-ready PyTorch layers verified across NVIDIA, AMD, and Intel GPUs.