Models that finish the job.
Less restrictive AI for coding, chat and research. Answers stream in real time, search the web with sources, and every model works through one API key.
Free plan included. No card needed to start.
My Node upload script runs out of memory on files over 2 GB. It reads the whole file into a Buffer and then calls s3.putObject. Fix it so memory stays flat.
Pick the model for the task
Six models, named by us, each tuned for a kind of work. Switch between them mid-conversation or call any of them from the API.
LS Lite
Fast everyday answers
- Context
- 128K tokens
- Speed
- Fast
- Pay as you go
- $0.20 in / $0.60 out per 1M tokens
LS Chat
Balanced general assistant
- Context
- 160K tokens
- Speed
- Balanced
- Pay as you go
- $0.50 in / $1.50 out per 1M tokens
LS Coder
Built for software engineering
- Context
- 256K tokens
- Speed
- Balanced
- Pay as you go
- $0.50 in / $2.00 out per 1M tokens
LS Thinker
Deep reasoning for hard problems
- Context
- 160K tokens
- Speed
- Deep
- Pay as you go
- $0.80 in / $3.50 out per 1M tokens
LS Scholar
Research with a 1M-token memory
- Context
- 1M tokens
- Speed
- Fast
- Pay as you go
- $0.50 in / $4.00 out per 1M tokens
LS Ultra
Our most capable model
- Context
- 200K tokens
- Speed
- Balanced
- Pay as you go
- $4.50 in / $22.50 out per 1M tokens
Three ways to pay, one account
Start on the free plan. Subscribe when you want bigger allowances. Add funds whenever you need a model your plan does not cover.
Free plan
A monthly token allowance on selected models, real-time chat, web searches and API keys. Allowances reset on the first of each month.
Paid plans
Each plan sets a monthly allowance per model, so you always know what is included. Pro and Max add the flagship model, more searches and monthly API credit.
Wallet, pay as you go
Add funds and pay per token for any model at the rate on its page. Set each API key to use your plan, your wallet, or the plan first and the wallet after.
Built for real work
Everything in the chat and the API is there because someone needed it to get a task done.
Real-time streaming
Tokens appear as they are generated in the chat and over the API. Stop a reply at any point and keep what was written.
Web search with sources
Turn on search for any conversation. The model reads current results and cites them so you can check every claim.
Context that follows you
Tell us who you are and how you like answers once. Save memories. Every model, every conversation, gets that context.
Models that know themselves
Each model knows its own name, context window and capabilities, and can recommend a better model for the task.
OpenAI-compatible API
Drop-in replacement for the OpenAI SDKs. Chat completions, streaming, JSON mode, models list, plus a web_search flag.
Usage you can audit
Every request is logged with tokens, cost and how it was billed. Per-model allowances show exactly what is left.
Plans
Prices in USD. Every plan can use the wallet for models it does not include.
Free
Try every workflow at no cost
$0 / month
- LS Lite and LS Chat with monthly allowances
- Real-time streaming chat
- 10 web searches a month
- Pay-as-you-go access to every model
- 2 models included, from 300K tokens a month
Pro
Most popularFor daily work and side projects
$20.00 / month
- Generous allowances on every model
- LS Ultra included
- 500 web searches a month
- $5 API credit every month
- 6 models included
Max
For heavy users and small teams
$60.00 / month
- 5x the Pro allowances
- LS Ultra at scale
- 2,000 web searches a month
- $15 API credit every month
- 6 models included
Full allowance table per model and pay-as-you-go rates on the pricing page.
Questions people ask before signing up
What does "less strict" actually mean?
Our models answer the ordinary requests other assistants refuse: mature fiction, security research, blunt feedback, medical and legal questions, dark humour, controversial topics. It is not a loophole for harm. Content that sexualises minors, enables mass-casualty weapons, attacks systems you do not own or defrauds people is banned and gets accounts closed. The full list is in the Acceptable Use Policy.
Are these your own models?
The names are ours; the weights are not. Each LessStrict model is a third-party model served through a provider we have selected for capability, speed and relaxed moderation. Every model page lists the capabilities, context window and prices, and we may swap the underlying provider to keep quality and price where we want them.
Do you train on my conversations?
No. Your messages are sent to the model provider that serves the selected model so it can answer, and stored in your history so you can come back to them. We do not use them to train models and we do not sell them.
Is the API compatible with the OpenAI SDK?
Yes. Point the official OpenAI Python or Node SDK at https://lessstrict.com/v1 with a LessStrict key and use our model slugs. Chat completions, streaming, JSON mode and the models list work as you expect, plus a web_search flag that is ours.
How do plan allowances and the wallet work together?
Every plan, including Free, includes a monthly token allowance per model. When a request fits inside the allowance it costs nothing extra. When a model is not in your plan or the allowance is used up, the request is charged per token from your prepaid wallet at the rate on the model page. You can also set a key or the chat to use only the plan or only the wallet.
What happens when I hit a limit?
You get a clear message naming the model, the limit and when it resets (daily limits at 00:00 UTC, monthly on the first). If you have wallet funds and your plan allows it, the request runs on pay-as-you-go instead. Nothing is charged for requests that fail before a model starts answering.
Can I get a refund?
First-time monthly subscribers can get a full refund within 7 days if they have used less than 20% of the allowance. Yearly plans are refundable pro rata within 30 days. Unused wallet top-ups can be refunded within 30 days minus payment fees. The Refund Policy has the details.
How do I cancel?
Dashboard, Billing, Cancel plan. Your plan stays active until the end of the period you paid for and does not renew. Your account, history and wallet balance stay as they are.
Start with the free plan today.
Create an account, open the chat and send the first message in under a minute.