DeepSeek: the right model and the right tool
Pricing, providers, ecosystem projects and self-hosting math — collected and verified, not hand-written. Every entry carries a date.
Picked from the ecosystem
From 4,150 collected projects, these are active and worth a look. Full data: ecosystem.json
DeepSeek Harness: Everything is a Plugin.
在 dsh 里装上这个插件即可，无需登录、注册或填 API Key，就能使用包括 DeepSeek V4.1 Flash、Kimi K3 在内的前沿模型——完全免费，不限量。 All you do is install
通过扫码或机器人凭据把IM机器人接入DeepSeek Harness（支持飞书、微信、钉钉、企业微信、QQ、Slack、Telegram、Discord和WhatsApp）。 Connect IM bots to Dee
免 ROOT 免 Termux，在手机上跑 DeepSeek Harness。完整 Ubuntu 环境 + proroot 零 ptrace 开销 · AI 输出实时上屏 · ADB 直连 · 数据不丢
免 ROOT 免 Termux，一个 APK 在安卓手机上跑 DeepSeek Harness。引擎内嵌 App 进程 · 读屏操作 App · 改系统设置 · 定时/条件自动化 · 可选 Shizuku
Open-source GrokBot alternative built on DeepSeek Harness. Build your personal bot team: multiple personal age
Run the DeepSeek Harness AI agent natively on Android — no root, no Termux, extension center included | 手机本地运行
Minimal desktop wrapper for DeepSeek Harness 极简 Deespeek Harness 桌面端，跨平台，免配置，开箱即用
Cheapest providers right now
DeepSeek V4.1 Flash 1017, sorted by input price. 6.0x cheaper than the official $0.15. Full quotes: provider comparison
| # | Provider | Input /1M | Output /1M | Quantization |
|---|---|---|---|---|
| 1 | Particle AI | $0.025 | $0.25 | n/d |
| 2 | Morph | $0.035 | $0.32 | FP8 |
| 3 | Relace | $0.035 | $0.50 | FP4 low |
| 4 | Wafer | $0.038 | $0.50 | mxfp4 |
| 5 | OpenInference | $0.038 | $0.50 | FP4 low |
| 6 | DeepInfra | $0.078 | $0.24 | FP8 |
Cheapest is not always best. FP4 and undisclosed
quantization can degrade reasoning and code output. Check the column before switching.
What do you need?
Run it on a device
Jetson, Raspberry Pi, ESP32, Home Assistant
Run it on a server
Ollama, llama.cpp, vLLM, quantization
Connect it to an app
LangChain, Dify, bots, Obsidian
Find a cheaper provider
197 quotes, up to 18x spread
Official pricing
Peak, off-peak, cache rates
Estimate your bill
Monthly cost from token volume
Official DeepSeek prices
| Model | Input /1M | Output /1M | Cache hit |
|---|---|---|---|
| Flash | $0.15 | $0.60 | $0.0030 |
| V4 Pro | $0.66 | $1.98 | $0.0220 |
Off-peak rates. Peak hours (Beijing 09:00–12:00, 14:00–18:00) cost double. Full table
Recently indexed
DeepSeek Harness: Everything is a Plugin.
Get up and running with Kimi, GLM, MiniMax, DeepSeek, gpt-oss, Qwen, Gemma and other models.
A cross-platform desktop All-in-One assistant for Claude Code, Codex, OpenCode,DeepSeek Harness and more. Only
✨ Zero-config AI chat assistant. No API key needed — sign up and instantly chat with GPT-5, Claude 4, Gemini 2
About
We collect DeepSeek ecosystem data from GitHub, Hugging Face, GitLab, npm and Docker Hub, score each entry for whether it actually works, and publish the result as pages and JSON. We are not affiliated with DeepSeek. We do not resell API access.
Last updated 2026-11-05.