MoE Models

MoE Models
MoE ModelsSpikingBrain: 100x Faster LLM Inference via Spike Sparsity
SpikingBrain leverages brain-inspired spike sparsity and Mixture of Experts (MoE) to drastically reduce LLM latency. It achieves a 100x TTFT speedup on 4M token sequences and supports MetaX or NVIDIA GPUs.
MoE Models
MoE ModelsDots.LLM1: 142B MoE Model Trained on 11.2T Real-World Tokens
Dots.LLM1 is a 142B total parameter MoE model with 14B activated. Trained on 11.2T tokens of real text, it rivals Qwen2.5-72B in performance at a lower computational cost.
MCP Services
MCP ServicesLiveMCPBench: Benchmark AI Agents on Real-World MCP Tool Tasks
LiveMCPBench evaluates AI agents on real-world tasks using 290+ MCP tools. Includes Office, Travel, Finance scenarios. Docker deployment guide included.
LLM Tooling
LLM ToolingAIPy: Execute Python via Natural Language Directly in Your Terminal
AIPy integrates Python with LLMs inside your terminal. Describe your task in plain English and let the tool write and execute the code. Streamline CSV cleaning, web scraping, and data analysis.
Data Processing Tools
Data Processing ToolsChinese Wikipedia Corpus: Processing 990k Articles for NLP Tasks
Learn how to clean and parse the 990,000-article Chinese Wikipedia dump. Utilize OpenCC for script conversion and a custom parser to avoid the common data loss associated with gensim's WikiCorpus.
PDF Tools
PDF ToolsPDF to Scanned Look Instantly – No Scanner Required
Convert PDFs into scanned-looking copies directly in your browser with LookScanned.io. No uploads, no printers, zero privacy risk – works fully offline as a PWA.
Liebao VPN
Liebao VPNLiebao VPN: Download, Install & Use on Android & iOS
Get Liebao VPN for Android, iOS, and smart TVs. Follow this comprehensive guide for installation, server switching, speed optimization, and advanced security features like 256-bit encryption.
1