What it does
5 features
Continue · Coding agent
Product and access
Available offering; access methods and prerequisites are described separately in this profile.
Core features
5 features from provider sources · reviewed on 26 October 2026. Availability and limitations follow below.
Usable AI models
Open choice
You add your models to config.yaml yourself, with your own key and a provider such as OpenAI, Anthropic or Mistral. OpenRouter gives you its model catalog, Ollama gives you local models, and apiBase also lets you use your own OpenAI-compatible endpoint. Agent mode requires a model that supports tools.
Model families named
- OpenAI (GPT)
- Anthropic (Claude)
- Mistral
- Other models can be connected freely
- Model choice
- You choose
- Your own API key
- Yes
- Router or custom endpoint
- Yes
- Local models
- Yes
The rating covers the generally available main access. Documented · Sources accessed on
Continue: config.yaml reference · Continue: Ollama · Continue: OpenRouter
Does it fit my task?
2 practical tasks · 3 requirements and limits
A good fit if …
- you want to automate coding tasks in the terminal or CI with cn -p
- you want to confirm write and Bash actions yourself before they run
- you want to sign in with a Continue account or your own Anthropic API key
Less suitable if …
- you need a managed solution with administrator approvals for a team; the reviewed pages describe none
- you need built-in schedules and failure notifications without your own scripts; the reviewed pages describe none
Editorial assessment based on provider documentation. Not a product test.
Practical tasks
Editorial examples based on documented features. No product tests performed.
Edit files
- Your input
- Read and edit project files.
- Possible result
- A work product for your review.
- What to check
- Review results and write actions before accepting them.
Run commands
- Your input
- Run commands for multi-step development tasks.
- Possible result
- A work product for your review.
- What to check
- Review results and write actions before accepting them.
Read and edit project files.
Your possible prompt
Read the project rules. Investigate [error], propose a small change, and run the relevant checks. Show me the diff and any remaining uncertainties.
Adjust the task, data and permissions.
Requirements and limits
Can I use it?
With requirements
Europe & regional availability
Consider installation and provider access separately. A complete regional feature matrix is not published.
Reviewed scope: Product access in Germany and the EU
Reviewed on:
With requirements
Access and connected services
Install the editor extension or CLI; the CLI requires a Continue account or Anthropic API key.
Region, reason and source
Region covered: Germany / EU
Documented reason: The reviewed sources document no specific restriction caused by the AI Act, DMA or GDPR. Model and service access depend on country and contract.
Official source · docs.continue.dev · Official source · docs.continue.dev
Accounts & permissions
Based on provider documentation. Permissions depend on the account, integration and organization policies. Review the permissions actually requested before use.
Configuration covered: Continue CLI (cn) on your own machine; the VS Code and JetBrains extension is only touched on
Sources reviewed:
Show all 3 points
Documented
Account used
You sign in to a Continue account in the browser with cn login, or enter an Anthropic API key on first launch. For scripts and CI you set the CONTINUE_API_KEY environment variable.
EvidenceOpen in reviewed sources
Admin approval
The reviewed pages describe no approval or restriction by organisation administrators. The tool permissions page covers local rules, flags and a permissions file.
EvidenceDocumented
Data access & team sharing
Read-only tools are allowed by default; write tools and Bash ask first. With --readonly (plan mode) write tools are excluded while reads and Bash stay allowed; with --auto everything runs without prompting. Rules can be narrowed with glob patterns on arguments such as file paths.
Evidence · Evidence
Memory and conversation continuity
3 of 6 criteria documented
Persistent rules and resumable chats; automatic fact memory is not established.
Continue IDE and cn CLI
Sources accessed:
Documentation by criterion
StoragePartly documented
Storage and scope
Configuration files and rules provide persistent instructions; earlier chats can be resumed. A separate learned fact store and its storage location are not described.
SelectionPartly documented
Selection and timing
Rules are added to system context; /init generates project instructions. Automatic conversation extraction into long-term memory is not documented.
Edit & deletePartly documented
Inspect, edit and delete
Rules are editable configuration; /clear clears chat. A separate memory list with fact editing is not documented.
Access boundariesDocumented
Agent and channel boundaries
Rules and context follow the selected agent configuration and project files. Chat resumption refers to a specific earlier session.
RetrievalDocumented
Retrieval
Rules are inserted into system messages; @ references load files. /resume restores conversation history; knowledge context is distinct from learned memory.
New conversationsDocumented
Session reset and compaction
/compact summarizes chat, /clear clears it and /resume continues an earlier chat. /fork branches the current history.
Cost and effort
Prices as of 26 October 2026
Prices and included usage
Pricing checked
Model costs depend on the selected access method. A current flat platform price is not confirmed by the reviewed sources.
- Model costs depend on the selected access method. A current flat platform price is not confirmed by the reviewed sources.
Setup
Install the editor extension or CLI; the CLI requires a Continue account or Anthropic API key.
Operation & oversight
Based on public provider documentation. Assessments are editorial inferences from the stated properties. Actual time requirements and reliability have not been measured.
Deployment covered: Continue CLI in the terminal (interactive) and in headless mode for scripts
Sources reviewed:
Editorial assessment
Control is yours: using --auto or --allow "*" removes the prompts, and plan mode still allows Bash. For unattended runs you must plan permissions, logging and retries yourself.
Show all 6 points
Open in reviewed sources
Sign-ins
The reviewed pages do not say how long a cn login sign-in or an API key remains valid.
EvidenceDocumented with conditions
Starting further runs
You start runs yourself: interactively with cn or as a single task with cn -p, for example in CI pipelines or git hooks. The headless page describes no built-in schedule.
EvidenceDocumented with conditions
Resuming after failures
With cn -p --resume you can resume a previous session without interaction; the earlier history is replayed. The headless page does not describe automatic retries.
Evidence
Evidence and open questions
2 sources · 1 open point
Open points
The reviewed sources contain no reliable information on these points. A gap says nothing about the product.
Requirements and limits: not yet confirmed
Sources and evidence
Official sources reviewed. Example tasks are editorial suggestions; hands-on testing is outstanding.
Editorial team: Michael Meierhoff · sources reviewed on . Provider information is not independent evidence of performance.
Provider
Coding agent for the editor and terminal with configurable models and tools.
- Operator
- Continue
- Organization
- Provider
- Website
- docs.continue.dev
Changes to this profile 1
Coding agent for the editor and terminal with configurable models and tools.
These notes describe editorial changes. Source and pricing review dates appear in their respective sections.