- claude-rolling-context — rolling context compression for Claude Code
- nestor-plugins — Claude Code plugin marketplace
- haste — agent harness
- Hugging Face — open weights
NodeNestor
Popular repositories Loading
-
claude-rolling-context
claude-rolling-context PublicRolling context compression for Claude Code — never hit the context wall. Auto-compresses old messages while keeping recent context verbatim. Zero config, zero latency. Works as a Claude Code plugin.
-
UE5UltimateMCP
UE5UltimateMCP PublicThe ultimate free AI integration for Unreal Engine 5 — 158 tools via Model Context Protocol
-
prism-optiscaler
prism-optiscaler PublicOptiScaler fork with Prism neural upscaler + FGExtrap depth-layered frame extrapolation
-
HiveMindDB
HiveMindDB PublicShared memory for AI agent swarms. Knowledge graphs, semantic search, LLM extraction, real-time channels — all Raft-replicated
Rust 6
Repositories
- paddock Public Forked from truespar/paddock
Native Rust inference server for open models on NVIDIA GPUs. OpenAI- and Anthropic-compatible APIs, GGUF + safetensors, FP8/NVFP4/MXFP4/Q8/Q4, built-in Studio
- trimode Public
Tri-mode (AR / block diffusion / self-speculative) conversion of Qwen3.5 GDN+attention hybrids - 0.8B experiment, aimed at the larger MoE hybrids
- claude-rolling-context Public
Rolling context compression for Claude Code — never hit the context wall. Auto-compresses old messages while keeping recent context verbatim. Zero config, zero latency. Works as a Claude Code plugin.
- claude-pii-proxy Public
Round-trip PII redaction proxy for Claude Code. Replaces names/emails/secrets with deterministic tokens before the API sees them; restores real values in the response. ~12ms/string.
- claude-model-router Public
Add custom models to Claude Code's /model picker. A local proxy routes each model to whatever backend you configure - local vLLM, llama.cpp, Ollama, LM Studio, OpenRouter - while everything else passes through untouched.
- haste Public
- claude-lean-context Public
Input-side token compression for Claude Code: read dedup-by-reference, codemap exploration reads, duplicate collapse. Context-aware companion to rolling-context.
- voicechat Public
A local voice assistant you talk to in a browser - audio-native listening via Gemma, one-file voice cloning via CSM, everything on your own GPU
Top languages
Loading…
Most used topics
Loading…