🐳 DeepSeek TUI

This terminal-native coding agent is built around DeepSeek V4's 1M-token context window and prefix cache capability. It is distributed as a single binary and requires no Node.js or Python runtime. It also includes an MCP client, a sandbox, and a durable task queue out of the box.

简体中文 README

Before proceeding, ensure that Node.js and npm are installed.

node --version
npm --version

If Node.js and npm are not installed in your environment, refer to the installation guides provided below:

• English version: https://nodejs.org/en/download
• Chinese version: https://nodejs.org/zh-cn/download

Select the version that best matches your device specifications and operating system.

Install the deepseek-tui now:

npm install -g deepseek-tui

# When installing deepseek-tui in China's internet environment, you can use an npm mirror to speed up the installation process.
# npm install -g deepseek-tui@latest --registry=https://registry.npmmirror.com

---

What Is It?

DeepSeek TUI is a coding agent that runs entirely in your terminal. It gives DeepSeek's frontier models direct access to your workspace — reading and editing files, running shell commands, searching the web, managing git, and orchestrating sub-agents — all through a fast, keyboard-driven TUI.

Built for DeepSeek V4 (deepseek-v4-pro / deepseek-v4-flash) with 1M-token context window and native thinking-mode (chain-of-thought) streaming.

Key Features

• Native RLM (rlm_query) — fans out 1–16 cheap deepseek-v4-flash children in parallel for batched analysis and parallel reasoning, all against the existing API client
• Thinking-mode streaming — watch the model's chain-of-thought unfold in real time as it works through your tasks
• Full tool suite — file ops, shell execution, git, web search/browse, apply-patch, sub-agents, MCP servers
• 1M-token context — automatic intelligent compaction when context fills up; prefix-cache aware for cost efficiency
• Three modes — Plan (read-only explore), Agent (interactive with approval), YOLO (auto-approved)
• Reasoning-effort tiers — cycle through off → high → max with Shift + Tab
• Session save/resume — checkpoint and resume long-running sessions
• Workspace rollback — side-git pre/post-turn snapshots with /restore and revert_turn, without touching your repo's .git
• Durable task queue — background tasks survive restarts; think scheduled automation, long-running reviews
• HTTP/SSE runtime API — deepseek serve --http for headless agent workflows
• MCP protocol — connect to Model Context Protocol servers for extended tooling; please see docs/MCP.md
• LSP diagnostics — inline error/warning surfacing after every edit via rust-analyzer, pyright, typescript-language-server, gopls, clangd
• User memory — optional persistent note file injected into the system prompt for cross-session preferences
• Localized UI — en, ja, zh-Hans, pt-BR with auto-detection
• Live cost tracking — per-turn and session-level token usage and cost estimates; cache hit/miss breakdown
• Skills system — composable, installable instruction packs from GitHub with no backend service required

---

How It's Wired

deepseek (dispatcher CLI) → deepseek-tui (companion binary) → ratatui interface ↔ async engine ↔ OpenAI-compatible streaming client. Tool calls route through a typed registry (shell, file ops, git, web, sub-agents, MCP, RLM) and results stream back into the transcript. The engine manages session state, turn tracking, the durable task queue, and an LSP subsystem that feeds post-edit diagnostics into the model's context before the next reasoning step.

See docs/ARCHITECTURE.md for the full walkthrough.

---

Quickstart

npm install -g deepseek-tui
deepseek --version
deepseek

Prebuilt binaries are published for Linux x64, Linux ARM64 (v0.8.8+), macOS x64, macOS ARM64, and Windows x64. For other targets (musl, riscv64, FreeBSD, etc.), see Install from source or docs/INSTALL.md.

On first launch you'll be prompted for your DeepSeek API key. The key is saved to ~/.deepseek/config.toml so it works from any directory without OS credential prompts.

You can also set it ahead of time:

deepseek auth set --provider deepseek   # saves to ~/.deepseek/config.toml

export DEEPSEEK_API_KEY="YOUR_KEY"      # env var alternative; use ~/.zshenv for non-interactive shells
deepseek

deepseek doctor                         # verify setup

To rotate or remove a saved key: deepseek auth clear --provider deepseek.

Linux ARM64 (Raspberry Pi, Asahi, Graviton, HarmonyOS PC)

npm i -g deepseek-tui works on glibc-based ARM64 Linux from v0.8.8 onward. You can also download prebuilt binaries from the Releases page and place them side by side on your PATH.

China / Mirror-friendly Installation

If GitHub or npm downloads are slow from mainland China, use a Cargo registry mirror:

# ~/.cargo/config.toml
[source.crates-io]
replace-with = "tuna"

[source.tuna]
registry = "sparse+https://mirrors.tuna.tsinghua.edu.cn/crates.io-index/"

Then install both binaries (the dispatcher delegates to the TUI at runtime):

cargo install deepseek-tui-cli --locked   # provides `deepseek`
cargo install deepseek-tui     --locked   # provides `deepseek-tui`
deepseek --version

Prebuilt binaries can also be downloaded from GitHub Releases. Use DEEPSEEK_TUI_RELEASE_BASE_URL for mirrored release assets.

Install from source

Works on any Tier-1 Rust target — including musl, riscv64, FreeBSD, and older ARM64 distros.

# Linux build deps (Debian/Ubuntu/RHEL):
#   sudo apt-get install -y build-essential pkg-config libdbus-1-dev
#   sudo dnf install -y gcc make pkgconf-pkg-config dbus-devel

git clone https://github.com/Hmbown/DeepSeek-TUI.git
cd DeepSeek-TUI

cargo install --path crates/cli --locked   # requires Rust 1.85+; provides `deepseek`
cargo install --path crates/tui --locked   # provides `deepseek-tui`

Both binaries are required. Cross-compilation and platform-specific notes: docs/INSTALL.md.

Other API Providers

# NVIDIA NIM
deepseek auth set --provider nvidia-nim --api-key "YOUR_NVIDIA_API_KEY"
deepseek --provider nvidia-nim

# Fireworks
deepseek auth set --provider fireworks --api-key "YOUR_FIREWORKS_API_KEY"
deepseek --provider fireworks --model deepseek-v4-pro

# Self-hosted SGLang
SGLANG_BASE_URL="http://localhost:30000/v1" deepseek --provider sglang --model deepseek-v4-flash

---

What's New In v0.8.11

A targeted patch for the V4 cache-maxing overhaul plus three runtime fixes discovered in YOLO long-session dogfooding. Full changelog.

• Cache-maxing prompt path for DeepSeek V4 — the engine skips system-prompt reassembly when the stable prefix is unchanged, moves the working-set summary out of the system prompt into per-turn metadata, and anchors the tool array with cache_control: ephemeral. Net effect: fewer prefix rewrites, higher cache-hit rates, lower cost per turn.
• 500K token compaction floor — automatic compaction refuses below 500K tokens; manual /compact still bypasses. The message-count trigger (a 128K-era heuristic) is removed — token budget is the sole auto-trigger.
• npm install resilience — retry with exponential backoff, per-attempt timeout + stall detector, HTTPS_PROXY/HTTP_PROXY/NO_PROXY support (pure Node, no new deps), and download progress to stderr. Driven by a community report from China where npm install took 18 minutes through a CN mirror.
• YOLO sandbox dropped — YOLO now uses DangerFullAccess (no sandbox), consistent with its auto-approval + trust mode posture. Previously, the WorkspaceWrite sandbox was intercepting legitimate outside-workspace writes.
• Scroll lock preserved during live streaming — scrolling up mid-stream no longer gets yanked back to the tail on the next chunk. The user_scrolled_during_stream lock now survives render-time clamping.
• Capacity controller off by default — the controller was silently clearing transcripts independent of token usage or auto_compact settings. Now defaults to disabled; opt-in via capacity.enabled = true.
• README clarity pass (#685) — title-cased headings, explicit prerequisites, China-friendly install variant. Thanks to @Agent-Skill-007 for this PR.

---

Usage

deepseek                                         # interactive TUI
deepseek "explain this function"                 # one-shot prompt
deepseek --model deepseek-v4-flash "summarize"   # model override
deepseek --yolo                                  # auto-approve tools
deepseek auth set --provider deepseek            # save API key
deepseek doctor                                  # check setup & connectivity
deepseek doctor --json                           # machine-readable diagnostics
deepseek setup --status                          # read-only setup status
deepseek setup --tools --plugins                 # scaffold tool/plugin dirs
deepseek models                                  # list live API models
deepseek sessions                                # list saved sessions
deepseek resume --last                           # resume latest session
deepseek serve --http                            # HTTP/SSE API server
deepseek pr <N>                                  # fetch PR and pre-seed review prompt
deepseek mcp list                                # list configured MCP servers
deepseek mcp validate                            # validate MCP config/connectivity
deepseek mcp-server                              # run dispatcher MCP stdio server

Keyboard Shortcuts

Key	Action
Tab	Complete / or @ entries; while running, queue draft as follow-up; otherwise cycle mode
Shift+Tab	Cycle reasoning-effort: off → high → max
F1	Searchable help overlay
Esc	Back / dismiss
Ctrl+K	Command palette
Ctrl+R	Resume an earlier session
Alt+R	Search prompt history and recover cleared drafts
Ctrl+S	Stash current draft (/stash list, /stash pop to recover)
@path	Attach file/directory context in composer
↑ (at composer start)	Select attachment row for removal
Alt+↑	Edit last queued message

Full shortcut catalog: docs/KEYBINDINGS.md.

---

Modes

Mode	Behavior
Plan 🔍	Read-only investigation — model explores and proposes a plan (update_plan + checklist_write) before making changes
Agent 🤖	Default interactive mode — multi-step tool use with approval gates; model outlines work via checklist_write
YOLO ⚡	Auto-approve all tools in a trusted workspace; still maintains plan and checklist for visibility

---

Configuration

User config: ~/.deepseek/config.toml. Project overlay: <workspace>/.deepseek/config.toml (denied: api_key, base_url, provider, mcp_config_path). config.example.toml has every option.

Key environment variables:

Variable	Purpose
DEEPSEEK_API_KEY	API key
DEEPSEEK_BASE_URL	API base URL
DEEPSEEK_MODEL	Default model
DEEPSEEK_PROVIDER	deepseek (default), nvidia-nim, fireworks, sglang
DEEPSEEK_PROFILE	Config profile name
DEEPSEEK_MEMORY	Set to on to enable user memory
NVIDIA_API_KEY / FIREWORKS_API_KEY / SGLANG_API_KEY	Provider auth
SGLANG_BASE_URL	Self-hosted SGLang endpoint
NO_ANIMATIONS=1	Force accessibility mode at startup
SSL_CERT_FILE	Custom CA bundle for corporate proxies

UI locale is separate from model language — set locale in settings.toml, use /config locale zh-Hans, or rely on LC_ALL/LANG. See docs/CONFIGURATION.md and docs/MCP.md.

---

Models & Pricing

Model	Context	Input (cache hit)	Input (cache miss)	Output
deepseek-v4-pro	1M	$0.003625 / 1M*	$0.435 / 1M*	$0.87 / 1M*
deepseek-v4-flash	1M	$0.0028 / 1M	$0.14 / 1M	$0.28 / 1M

Legacy aliases deepseek-chat / deepseek-reasoner map to deepseek-v4-flash. NVIDIA NIM variants use your NVIDIA account terms.

DeepSeek Pro rates currently reflect a limited-time 75% discount, which remains valid until 15:59 UTC on 5 May 2026. After that time, the TUI cost estimator will revert to the base Pro rates.

---

Publishing Your Own Skill

DeepSeek TUI discovers skills from workspace directories (.agents/skills → skills → .opencode/skills → .claude/skills) and the global ~/.deepseek/skills. Each skill is a directory with a SKILL.md file:

~/.deepseek/skills/my-skill/
└── SKILL.md

Frontmatter required:

---
name: my-skill
description: Use this when DeepSeek should follow my custom workflow.
---

# My Skill
Instructions for the agent go here.

Commands: /skills (list), /skill <name> (activate), /skill new (scaffold), /skill install github:<owner>/<repo> (community), /skill update / uninstall / trust. Community installs from GitHub require no backend service. Installed skills appear in the model-visible session context; the agent can auto-select relevant skills via the load_skill tool when your task matches their descriptions.

---

Documentation

Doc	Topic
ARCHITECTURE.md	Codebase internals
CONFIGURATION.md	Full config reference
MODES.md	Plan / Agent / YOLO modes
MCP.md	Model Context Protocol integration
RUNTIME_API.md	HTTP/SSE API server
INSTALL.md	Platform-specific install guide
MEMORY.md	User memory feature guide
SUBAGENTS.md	Sub-agent role taxonomy and lifecycle
KEYBINDINGS.md	Full shortcut catalog
RELEASE_RUNBOOK.md	Release process
OPERATIONS_RUNBOOK.md	Ops & recovery

Full Changelog: CHANGELOG.md.

---

Thanks

Earlier releases shipped with help from these contributors:

• Hafeez Pizofreude — SSRF protection in fetch_url and Star History chart
• Unic (YuniqueUnic) — Schema-driven config UI (TUI + web)
• Jason — SSRF security hardening

---

Contributing

See CONTRIBUTING.md. Pull requests welcome — check the open issues for good first contributions.

Note

Not affiliated with DeepSeek Inc.

License

MIT

Star History