01
Feature access
Pro and Max include Canvas and Kode access. Opening, editing or organising work in those tools does not spend credits.
Pricing
Every plan starts with K-AI 1.0. Compare exact model requests, context, imports, storage, search, creative tools and AI usage credits before you choose.
Monthly or annual
Plus is the best fit for most people. Pro and Max add Live, Canvas, Kode and Frontier models for more demanding work.
Best for trying Kontinue AI
Explore K-AI and import your first conversations.
No payment required
Start freePlan highlights
Best for casual AI users
Affordable access to basic models for everyday questions.
Billed monthly
Get StarterPlan highlights
Best for regular AI users
Use multiple models daily with more requests and imports.
Billed monthly
Get PlusPlan highlights
How credits work
AI usage credits keep different models and tools on one understandable balance. They do not charge you merely for opening a feature.
01
Pro and Max include Canvas and Kode access. Opening, editing or organising work in those tools does not spend credits.
02
Paid-model requests, web search, K-Image, K-Video, Live, Canvas AI actions and Kode AI runs use the shared credit balance.
03
Each plan has request limits and a credit balance. Costlier model groups and heavier tools use more credits than Basic requests.
Basic request
1 credit
Pro request
6 credits
Frontier request
25 credits
Web search
4 credits
K-Image
20 credits
K-Video
from 15/second
Kontinue Live
2/second
Kode AI unit (up to 25K tokens)
25 credits
These are operation rates. The feature-specific request, duration and build ceilings in the comparison table still apply.
Feature comparison
Values are shown wherever quantity matters. On smaller screens, pick a plan and read its limits one by one.
K-AI requests
Monthly requests to Kontinue AI’s native intelligence layer.
Basic-model requests
Cost-efficient models for everyday writing, analysis, coding and multimodal work. Requests also draw from AI usage credits.
Pro-model requests
Higher-capability models such as supported GPT, Claude Sonnet, Gemini Pro and Perplexity options. Requests also draw from AI usage credits.
Frontier-model requests
Our most capable included models, with tighter context and request limits because their provider costs are higher.
AI usage credits
Shared metering for paid chat tokens and K-AI fallbacks, web search, K-Image, K-Video, Kontinue Live, AI actions in Canvas, and Kode AI runs. Simply opening Canvas or Kode does not use credits.
Included model access
The exact catalogue can change as providers release, reprice or retire models. Kontinue shows only the models available to your plan.
K-AI context / output
Context includes system instructions, conversation history, files, tools and retrieved memory. Output is the maximum generated response.
Basic context / output
Maximum total input context and generated output for Basic models.
Pro context / output
Maximum total input context and generated output for Pro models.
Frontier context / output
Maximum total input context and generated output for Frontier models.
Conversation imports
Imports created from supported shared links or data exports. Fair-use soft caps protect the service from automated bulk abuse.
Data-import archive size
Maximum size of a supported account or conversation archive. This is separate from chat file uploads.
Chat file uploads
Files attached directly to conversations for analysis. Daily limits reset at 00:00 UTC.
Maximum chat file size
Stored files
Storage for uploaded documents and media. This is separate from searchable AI memory.
Export and import your data
Download supported Kontinue data and import supported exports without losing ownership of your information.
Web searches
Provider-backed searches that retrieve current web information. A search also consumes AI usage credits.
Search availability
K-Image generations
Monthly safety cap for K-Image. Each generation also consumes AI usage credits based on provider cost.
K-Video generation
Maximum generated video duration per month. Video generation also consumes AI usage credits and varies by resolution and quality.
AI memory
Searchable memories Kontinue can retrieve to personalise future work. This is not ordinary file storage.
Kontinue Live
Realtime voice sessions. Session time and the underlying realtime model usage both count toward your plan limits.
Canvas
The Canvas workspace is included on Pro and Max. Opening and editing Canvas does not consume credits; AI generation inside it does.
Kode
Kode access and build-count limits are included. Kode reserves AI units while it works and returns unused units. One unit covers up to 25K model tokens and costs 25 shared AI usage credits.
| Features | Free $0 forever | Starter $5.99/month $5/month annually | PlusPopular $19.99/month $16.99/month annually | Pro $59.99/month $56.99/month annually | Max $149.99/month $146.99/month annually |
|---|---|---|---|---|---|
| AI and models | |||||
| K-AI requests | 1,000/month | 2,000/month | 3,000/month | 5,000/month | 10,000/month |
| Basic-model requests | — | 100/month | 800/month | 2,000/month | 4,000/month |
| Pro-model requests | — | — | 100/month | 300/month | 700/month |
| Frontier-model requests | — | — | — | 25/month | 100/month |
| AI usage credits | 250 | 1,000 | 4,000 | 20,000 | 45,000 |
| Included model access | K-AI | K-AI + 3 Basic models | K-AI + Basic + Pro | K-AI + Basic + Pro + Frontier | All included model groups |
| Context per request | |||||
| K-AI context / output | 8K / 1K tokens | 16K / 2K tokens | 32K / 4K tokens | 64K / 8K tokens | 128K / 16K tokens |
| Basic context / output | — | 8K / 1K tokens | 16K / 2K tokens | 32K / 4K tokens | 128K / 8K tokens |
| Pro context / output | — | — | 8K / 2K tokens | 16K / 4K tokens | 64K / 8K tokens |
| Frontier context / output | — | — | — | 8K / 4K tokens | 32K / 8K tokens |
| Conversations and files | |||||
| Conversation imports | 10/month | 50/month | 1,000/month | Unlimited fair use (5,000 soft cap) | Unlimited fair use (20,000 soft cap) |
| Data-import archive size | 500 MB | 500 MB | 1 GB | 2 GB | 2 GB |
| Chat file uploads | — | 3/day | 10/day | 50/day | 200/day |
| Maximum chat file size | — | 5 MB | 20 MB | 50 MB | 100 MB |
| Stored files | — | 250 MB | 2 GB | 10 GB | 100 GB |
| Export and import your data | |||||
| Search and creation | |||||
| Web searches | 30/month | 100/month | 300/month | 1,000/month | 3,000/month |
| Search availability | K-AI only | K-AI only | All available models | All available models | All available models |
| K-Image generations | — | 10/month | 50/month | 200/month | 500/month |
| K-Video generation | — | — | — | 60 sec/month | 180 sec/month |
| Memory and professional tools | |||||
| AI memory | 0.5 MB | 5 MB | 100 MB | 500 MB | 2 GB |
| Kontinue Live | — | — | — | 30 min/month | 120 min/month |
| Canvas | Not included | Not included | Not included | ||
| Kode | — | — | — | Kode Lite · 25 builds/month | Full Kode · 100 builds/month |
Request counts are monthly safety ceilings. AI usage credits are shared across metered AI work, so higher-cost model groups, media, Live and Kode use the balance faster. Limits reset monthly at 00:00 UTC.