Deepseek V4 Python Packages

mcore-bridge

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-5.1, Deepseek-V4, MiniMax-2.7, ...) and 200+ multimodal large models (Qwen3.5, Qwen3-Omni, Gemma4, ...).

nemo-automodel

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

deepseek-harness

Harness for DeepSeek V4-Pro / V4-Flash. Python lib (pip install deepseek-harness) + dsh CLI + MCP server (npx @deepseek-harness/mcp) + Anthropic SKILL.md. 16 documented protocol quirks, 12 probes, 270+ trials.

deepseek-harness-cli

Harness for DeepSeek V4-Pro / V4-Flash. Python lib (pip install deepseek-harness) + dsh CLI + MCP server (npx @deepseek-harness/mcp) + Anthropic SKILL.md. 16 documented protocol quirks, 12 probes, 270+ trials.

deepseek-1m

🚀 Chat with your entire codebase, book, or email archive using DeepSeek-V4's 1M token context. No RAG. No chunking. Just pure memory.