deepseek-v4

🚀 Pytorch Distributed native training library for LLMs/VLMs with OOTB Hugging Face support

更适配中文用户的轻量开源AI Agent | 国产大模型Coding plan支持 | 兼容OpenClaw Skills生态| 已接入微信ClawBot/微博龙虾/飞书/钉钉/QQ/小智AI/Telegram/deepseek-v4。

Local-first AI agent runtime with sandboxed sessions, MCP tools, memory, credentials, audit/replay, and a built-in console. Run OpenAI, Anthropic, MiniMax, DeepSeek V4, and OpenAI-compatible models on your infrastructure.

A native .NET LLM inference engine for GGUF models. TensorSharp provides a console application, a web-based chatbot interface, and Ollama/OpenAI-compatible HTTP APIs for programmatic access. It supports Windows/MacOS/Linux with full GPU capability

Go SDK for DeepSeek V4 API — chat, reasoning, Anthropic-compatible endpoint, tool calling, FIM, and streaming. Also supports OpenRouter, Azure, Ollama, and OpenAI-compatible providers.

MCore-Bridge: Providing Megatron-Core model definitions for state-of-the-art large models and making Megatron training as simple as Transformers — with support for 300+ large language models (Qwen3-Next, GLM-5.2, Deepseek-V4, MiniMax-2.7, ...) and 200+ multimodal large models (Qwen3.5, Qwen3-Omni, Gemma4, ...).

Reproducible refusal-subspace editing and TP=2 deployment of DeepSeek V4 Flash 0731 on two DGX Sparks

Two-node DGX Spark/ASUS GX10 DeepSeek V4 Flash DSpark NVFP4 port for vLLM 0.25, with live dashboard and reproducible deployment.

Python-side witness stack for the official DeepSeek Harness (Node). Five `dsh doctor --node` probes each report something the Node runtime will not. `pip install deepseek-harness-cli` first

CLI & async Python library for free AI chat, image & video generation.