02
What each one costs
List prices for all 27 tools, checked 3 August 2026. The Continuum app itself is $0 on every row - hosted inference is a separate, optional plan.
checked 3 aug 2026
Tool
What it is
List price
Why people add Continuum
Conductor
Parallel agent orchestrator
Free · $50 Pro · $60/user Teams
Same worktrees, plus phone approve and spend by repo
Cursor
AI-first IDE
$0 / $20 / $60 / $200 · $40 seat
Keep the editor; run the long agents outside it
Claude Code
Official Anthropic coding CLI
$20 / $100 / $200 (Anthropic)
Your Claude plan, with gauges and multi-account
Codex
OpenAI coding agent
$8 / $20 / $200 (OpenAI)
Your ChatGPT plan, beside Claude in one sidebar
Windsurf (Devin Desktop)
AI IDE (Cognition)
$0 / $20 / $200 · $80 + $40 seat
Cascade stays; agent fleet moves to Continuum
GitHub Copilot
AI pair programmer
$0 / $10 / $39 / $100 · $19 seat
Completions stay; agent runs get a control plane
Devin
Autonomous software engineer
Free · $20 Pro · $200 Max · teams
Supervised agents on your hosts, not a service
Aider
CLI pair programming agent
Free OSS + your tokens
Same terminal habit, plus mobile and quota gauges
OpenCode
Open multi-provider coding agent
Free MIT · Go $10/mo
OSS agent with Code panes and multi-account
Gemini CLI / Antigravity
Google coding agents
Antigravity free tier or API key
Google agents beside Claude and Codex
Hermes Agent
Self-improving autonomous agent
Free MIT + credits
Research loops stay; shipping work gets panes
OpenClaw
Self-hosted personal AI assistant
Free MIT + hosting
Assistant stays; repo work gets worktrees
Grok Bot
Managed always-on cloud teammates (SpaceXAI)
$40-$300 plan + uncapped meter
Cloud teammates stay; repo work runs on your own hardware
Pi Coding Agent
Minimal coding agent harness
Free MIT + your tokens
Minimal harness, plus a product surface when needed
Factory AI
Enterprise autonomy stack
$20 / $100 / $200 · no free tier
Your own plans instead of a platform seat
T3 Code
Open-source coding-agent workbench
Free MIT + your subscriptions
An MIT agent control plane you can fork and self-host. Continuum answers with one polished live session across every screen.
Augment Code
Enterprise context-engine agent platform
$100/mo Business · Enterprise custom
Deep organizational context and managed agent loops, versus a free cockpit for the vendor agents you already run.
Qodo
AI code review and governance
$30 Pro Team · Enterprise custom
Continuum runs the agent fleet. Qodo reviews what that fleet writes and enforces the rules around it.
Morph LLM
Inference infrastructure
Free credits · usage based · $20 Starter
Morph makes agent infrastructure fast. Continuum gives the person operating agents a workbench.
OpenRouter
Model aggregation gateway
Credits · per token · 5.5% platform fee
OpenRouter sells access to 500+ models per token. Continuum sells a coding workbench with a flat weekly hosted allowance.
LiteLLM
Open-source LLM proxy and SDK
Free open source · Enterprise custom
LiteLLM is the proxy you run. Continuum is the outcome that proxy is usually built to deliver for coding agents.
Requesty
AI routing gateway
Free tier · 5% markup · Enterprise custom
Requesty routes 600+ models at a 5% markup. Continuum runs the coding agents and prices the work by repository.
Helicone
LLM observability and gateway
Free Hobby · $79 Pro · $799 Team
Helicone watches your LLM app in production. Continuum watches and controls what coding agents cost across a team.
Portkey
Enterprise AI gateway
Free OSS · $49/mo · Enterprise custom
Portkey governs the LLM traffic your apps send through a proxy. Continuum governs the coding agents running on developer machines.
TrueFoundry
Enterprise AI gateway and LLMOps platform
Free 50k · $499 Pro · $2,999 Pro Plus
TrueFoundry is a Kubernetes-native platform for running gateways and models in your own VPC. Continuum governs coding agents without any of that.
Fireworks
Open-model inference and training platform
$1 credits · usage based · GPUs $7/hr
Fireworks serves open models fast and meters every token. Continuum is the workbench those models get used in, at a flat monthly fee.
Together
Open-model inference and GPU cloud
Usage based · batch 50% off · H100 $3.99
Together sells breadth and raw GPUs by the hour. Continuum is the workbench that consumes them, with a flat-fee option instead of a meter.
Vendors reprice often. Every figure links through to the vendor's own pricing page on the matchup, and each page states the date it was checked.