Terminal Coding Agents
This category focuses on products with a usable terminal workflow, not products that only expose chat inside a browser or editor.
Selection criteria
- Provides a documented terminal or command-line interface
- Can work with source code or repositories
- Supports more than simple one-shot code completion
Editorial layer
Featured entities
Featured entries are curated separately from the automatic graph aggregation below.
Claude Code
Claude Code is Anthropic's coding agent for working directly with a codebase from the terminal, supported IDEs, the web, GitHub, Slack, and other product surfaces.
Codex CLI
Codex CLI is OpenAI's open-source coding agent that runs locally in a terminal. It can inspect a repository, edit files, and run commands under configurable approval controls.
Gemini CLI
Gemini CLI is Google's open-source terminal agent for working with Gemini models. It includes file operations, shell commands, web fetching, Google Search grounding, and MCP support.
Qwen Code
Qwen Code is an open-source coding agent from the Qwen team with terminal, editor, desktop, browser, and chat surfaces and support for Qwen and third-party model providers.
Aider
Aider is an open-source terminal tool for working with language models on an existing codebase. It builds a repository map, edits files, integrates with Git, and can run linting and tests.
Direct members
Tools
Aider
Aider is an open-source terminal tool for working with language models on an existing codebase. It builds a repository map, edits files, integrates with Git, and can run linting and tests.
Amazon Q Developer
Amazon Q Developer is an AWS coding and cloud assistant available in IDEs, on the command line, in AWS applications, and through other supported integrations.
Cursor
Cursor is a code editor from Anysphere with code completion and agents that can inspect repositories, edit files, run terminal commands, and work from desktop, terminal, or web surfaces.
GitHub Copilot
GitHub Copilot provides code completion, chat, agent mode, code review, a cloud coding agent, and a command-line interface across GitHub and supported development environments.
Kiro
Kiro is an AWS development environment with a desktop IDE, command-line agent, web interface, mobile access, and shared agent configuration across those surfaces.
Tabnine
Tabnine is an enterprise AI coding platform with IDE and CLI workflows, code completion, chat, agents, code review, organizational context, and deployment options including SaaS, VPC, on-premises, and air-gapped environments.
Direct members
Agents
Amp
Amp is a coding agent from Amp Frontier Corporation that runs in the terminal, on the web, in cloud orbs, and through desktop and mobile apps, with integrations for common code editors.
Claude Code
Claude Code is Anthropic's coding agent for working directly with a codebase from the terminal, supported IDEs, the web, GitHub, Slack, and other product surfaces.
Cline
Cline is an open-source coding agent available as an IDE extension, command-line tool, and SDK. It can read and write files, run terminal commands, and use browser and external tools.
Codex CLI
Codex CLI is OpenAI's open-source coding agent that runs locally in a terminal. It can inspect a repository, edit files, and run commands under configurable approval controls.
Devin CLI
Devin CLI is Cognition's local coding agent for terminal workflows. It can work with multiple model providers, run against a local repository, and hand sessions off to Devin Cloud.
Droid
Droid is Factory's software-development agent for planning, implementing, reviewing, and validating code changes across desktop, terminal, web, and managed execution environments.
Gemini CLI
Gemini CLI is Google's open-source terminal agent for working with Gemini models. It includes file operations, shell commands, web fetching, Google Search grounding, and MCP support.
Junie
Junie is JetBrains' coding agent for JetBrains IDEs and the command line. It can inspect projects, write and modify code, run development tasks, and use project instructions and agent skills.
OpenCode
OpenCode is an open-source coding agent available through a terminal interface, desktop app, and IDE integrations, with support for multiple model providers and local models.
OpenHands
OpenHands is an open-source platform for running software-development agents locally, in hosted cloud environments, or through self-hosted deployments.
Pi
Pi is an open-source, extensible terminal coding agent from Earendil Works with multi-provider model access, local tools, skills, extensions, prompt templates, and SDK support.
Qwen Code
Qwen Code is an open-source coding agent from the Qwen team with terminal, editor, desktop, browser, and chat surfaces and support for Qwen and third-party model providers.
Graph expansion
Models and APIs
Includes indexed models and APIs directly assigned to the category or connected to its tools and agents by verified usage, integration, power, or availability relationships.
Models
Claude Fable 5.1
Claude Fable 5.1 is an Anthropic model for demanding reasoning and long-horizon agentic work, with a one-million-token context window and up to 128K output tokens.
Claude Opus 5.5
Claude Opus 5.5 is Anthropic's current Opus model for long-running agentic coding and knowledge work, with a 1M-token context window.
Claude Sonnet 5.5
Claude Sonnet 5.5 is an Anthropic model released in September 2026 with a one-million-token context window, text and image input, and text output.
DeepSeek V4 Pro
DeepSeek V4 Pro is a reasoning and agent-oriented model in the DeepSeek V4 family, available through the DeepSeek API with a one-million-token context window.
Gemini 3.8 Flash
Gemini 3.8 Flash is a Google model for long-horizon software engineering, agent workflows, and multimodal tasks, available through the Gemini API.
GPT-5.6 Luna
GPT-5.6 Luna is OpenAI's cost-optimized GPT-5.6 model for high-volume workloads, with a 1.05-million-token context window and up to 128K output tokens.
GPT-5.6 Sol
GPT-5.6 Sol is OpenAI's flagship GPT-5.6 model for complex professional work, with a 1.05-million-token context window and up to 128K output tokens.
GPT-5.6 Terra
GPT-5.6 Terra is an OpenAI model designed to balance capability and cost, with a 1.05-million-token context window and up to 128K output tokens.
GPT-6 Astra
GPT-6 Astra is OpenAI's flagship API model for complex reasoning, coding, computer use, research, and other tool-heavy professional workflows.
GPT-6.1 Sol
GPT-6.1 Sol is an OpenAI model positioned below Astra on price while retaining support for complex coding, computer use, reasoning, and tool-based workflows.
Grok 4.7
Grok 4.7 is SpaceXAI's current flagship model for coding and general agentic work, with a 500K-token context window and configurable reasoning.
Qwen3-Coder-Plus
Qwen3-Coder-Plus is a Qwen coding model served through Alibaba Cloud Model Studio and included in the Alibaba Cloud Coding Plan configuration for Qwen Code.
APIs
Alibaba Cloud Model Studio API
Alibaba Cloud Model Studio API provides programmatic access to Qwen and supported third-party models through native DashScope and OpenAI-compatible interfaces.
Claude API
Claude API is Anthropic's REST API for programmatic access to Claude models, including Messages, streaming, tool use, files, model listing, and related platform features.
DeepSeek API
DeepSeek API provides access to the current DeepSeek V4 models through OpenAI-compatible and Anthropic-compatible interfaces.
Gemini API
Gemini API provides programmatic access to Google's Gemini and related generative models through standard, streaming, real-time, batch, embeddings, and media endpoints.
Groq API
Groq API is an OpenAI-compatible inference API for supported language and audio models, with Responses, Chat Completions, streaming, tool use, structured output, and speech endpoints.
Mistral API
Mistral API provides programmatic access to Mistral models, chat completions, agents, tool use, structured outputs, embeddings, document workflows, and batch processing.
OpenAI API
OpenAI API provides programmatic access to OpenAI models through the Responses API, Realtime API, batch processing, and other model and media endpoints.
OpenRouter API
OpenRouter API is an OpenAI-compatible model-routing API that exposes hundreds of models from many providers through a single endpoint and account.
xAI API
xAI API provides programmatic access to Grok models through OpenAI-compatible REST endpoints and xAI's gRPC interface, with text, tool, media, file, and batch APIs.
Provider graph
Providers
Organizations linked to the category's tools, agents, models, or APIs through verified develops, provides, operates, or owns relationships.
Alibaba Cloud
Alibaba Cloud operates Model Studio, a managed model platform that provides Qwen and third-party models through native and compatible APIs.
All Hands AI
All Hands AI is the company behind the OpenHands open-source software-development agent platform.
Amazon Web Services
Amazon Web Services is Amazon's cloud business and provides cloud, developer, and AI services including Amazon Q Developer.
Amp Frontier Corporation
Amp Frontier Corporation develops Amp. The Amp team announced in December 2025 that it was spinning out of Sourcegraph as an independent company.
Anomaly
Anomaly is the organization behind OpenCode and related open-source developer infrastructure projects including models.dev.
Anthropic
Anthropic is an AI research and product company that develops Claude models, the Claude API, and products including Claude Code.
Anysphere
Anysphere develops Cursor, a software platform and code editor focused on coding workflows and coding agents.
Cline Bot
Cline Bot develops Cline, an open-source coding agent available through IDE, command-line, and SDK interfaces.
Cognition
Cognition develops the Devin product family for local, desktop, and cloud software-engineering workflows and acquired Windsurf in 2025.
DeepSeek
DeepSeek develops language models, an API platform, and agent-oriented model tooling, including hosted DeepSeek V4 models and open-weight DeepSeek V3 and R1 research families.
Earendil Works
Earendil Works is an open-source software organization whose public projects include the Pi agent toolkit and coding-agent CLI.
Factory
Factory develops Droid and a software-development platform for running, governing, and automating agent-based engineering workflows.
GitHub
GitHub operates a software development platform for hosting repositories, collaboration, automation, code review, and developer tools.
Google develops internet, cloud, developer, and AI products, including the Gemini model and developer-tool product family.
Groq
Groq operates an inference cloud and API for running supported language, audio, and agent-oriented models on Groq infrastructure.
JetBrains
JetBrains develops developer tools and IDEs, including the Junie coding agent and JetBrains AI products.
Mistral AI
Mistral AI develops language models and operates Studio, its developer platform for model APIs, agents, document processing, and related AI workloads.
OpenAI
OpenAI is an AI research and deployment company that develops model, developer, and coding products including the Codex product family.
OpenRouter
OpenRouter operates a model-routing platform and unified API for accessing models from multiple AI providers through one account and endpoint.
SpaceXAI
SpaceXAI develops the Grok model family and operates the xAI API. xAI joined SpaceX in February 2026.
Tricentis
Tricentis develops enterprise software quality and testing products and acquired the Tabnine AI coding platform in 2026.
Research layer
Comparisons
Published comparisons are included when at least one compared entity is a direct member of this category or a model/API reached through its verified graph connections.
Aider vs Cline
Aider and Cline are open-source coding tools that can edit repositories and work with multiple model providers. Aider is terminal-centered with repository-map and Git workflows, while Cline spans IDE, CLI, and desktop surfaces with browser and MCP tooling.
Amp vs Claude Code
Amp and Claude Code are coding agents with terminal workflows. Amp uses multi-model routing and can run work locally or in cloud orbs; Claude Code is Anthropic's coding agent built around Claude models and Anthropic-managed product surfaces.
Claude Code vs Codex CLI
Claude Code and Codex CLI are coding agents with local terminal workflows. Claude Code also has Anthropic-managed IDE and web surfaces, while Codex CLI is the open-source local command-line client in the Codex family.
Claude Code vs Cursor
Claude Code is Anthropic's coding agent across terminal, IDE, web, and managed integrations. Cursor is a standalone code editor with local and cloud agent workflows plus CLI and web access.
Cursor vs GitHub Copilot
Cursor is a standalone editor from Anysphere. GitHub Copilot runs across GitHub, supported IDEs, and the command line. Both include agent workflows.
DeepSeek API vs OpenAI API
DeepSeek API and OpenAI API both support the OpenAI Responses request style. DeepSeek additionally documents an Anthropic-compatible base URL, while OpenAI exposes a broader first-party set of realtime and media endpoints.
DeepSeek V4 Pro vs GPT-6.1 Sol
DeepSeek V4 Pro and GPT-6.1 Sol are current reasoning models used in coding and agent workflows. DeepSeek documents a larger maximum output, while GPT-6.1 Sol accepts image input and has a slightly larger context window.
Devin Desktop vs Cursor
Devin Desktop, formerly Windsurf, combines an IDE with a command center for local and cloud agents. Cursor is an AI code editor with integrated agent and cloud-agent workflows.
Droid vs Devin
Droid and Devin are software-development agents that can carry out multi-step engineering work. Droid is exposed across local, desktop, CLI, web, and managed cloud surfaces; Devin is centered on Cognition-managed cloud environments.
Gemini 3.8 Flash vs GPT-6.1 Sol
Gemini 3.8 Flash and GPT-6.1 Sol both target coding and agent workflows. Gemini 3.8 Flash accepts a broader set of native media inputs, while GPT-6.1 Sol has a larger documented maximum output.
Gemini API vs OpenAI API
Gemini API and OpenAI API both expose multimodal models, streaming, tool use, batch processing, and official SDKs. Their primary request primitives and authentication schemes differ.
Gemini CLI vs Codex CLI
Gemini CLI and Codex CLI are Apache-2.0 terminal coding agents. Gemini CLI is centered on Google's Gemini ecosystem and includes built-in web tools, while Codex CLI is OpenAI's local Codex client with configurable approval controls.
GitHub Copilot vs Junie
GitHub Copilot is a coding assistant across GitHub, supported IDEs, and the command line. Junie is JetBrains' coding agent for JetBrains IDEs and the command line, with project instructions, skills, MCP, and custom model access.
GPT-6.1 Sol vs Claude Sonnet 5.5
GPT-6.1 Sol and Claude Sonnet 5.5 are current general-purpose models with strong coding and agent-workflow support. Their standard API input/output prices are both USD 2 and USD 10 per million tokens.
GPT-6 Astra vs Claude Opus 5.5
GPT-6 Astra and Claude Opus 5.5 are high-capability models from OpenAI and Anthropic. Both support text and image input, reasoning, tool-based workflows, and long contexts.
Grok 4.7 vs GPT-6.1 Sol
Grok 4.7 and GPT-6.1 Sol are current multimodal models used for coding and agent workflows. Grok 4.7 has lower standard output-token pricing, while GPT-6.1 Sol has a larger documented context window.
Groq API vs OpenRouter API
Groq API and OpenRouter API both expose OpenAI-compatible model access. Groq runs a selected model catalog on its inference cloud, while OpenRouter routes requests across many model providers.
Kiro vs Cursor
Kiro and Cursor both combine code editing with agent workflows. Kiro spans IDE, CLI, web, and mobile surfaces, while Cursor centers on its editor with local and cloud agent workflows.
Mistral Medium 3.5 vs Claude Sonnet 5.5
Mistral Medium 3.5 and Claude Sonnet 5.5 both target coding and agentic work. Mistral Medium 3.5 is released as open weights, while Claude Sonnet 5.5 is served as a proprietary Anthropic model.
OpenAI API vs Claude API
OpenAI API and Claude API both provide model access, streaming, tool use, and official SDKs. They differ in API surface, model catalog, authentication details, and supported media and realtime endpoints.
OpenCode vs Claude Code
OpenCode is an open-source, provider-agnostic coding agent with terminal, desktop, and IDE surfaces. Claude Code is Anthropic's coding agent with terminal, IDE, web, and other Claude product integrations.
OpenHands vs Devin
OpenHands and Devin both target multi-step software-development work. OpenHands is an MIT-licensed platform that can run locally, in its hosted service, or self-hosted; Devin is a proprietary cloud agent from Cognition.
OpenRouter API vs OpenAI API
OpenRouter API is a multi-provider routing layer with an OpenAI-compatible interface, while OpenAI API is OpenAI's first-party platform for its own models and developer services.
Qwen Code vs Gemini CLI
Qwen Code and Gemini CLI are Apache-2.0 coding agents that run from the terminal. Qwen Code supports multiple provider protocols and additional desktop, browser, and editor surfaces; Gemini CLI is centered on Google's Gemini ecosystem.
xAI API vs OpenAI API
xAI API and OpenAI API both expose Responses-style and OpenAI-compatible developer workflows, with different model catalogs, server-side tools, media endpoints, and transport options.
Zed vs Cursor
Zed is an open-source editor with native, external, and terminal-backed agent paths. Cursor is a proprietary editor centered on its own agent workflows and cloud agents.