Simple, transparent pricing
Every plan includes dedicated server compute and access to all AI models. Pay only for what you use with our credit system.
Starter
24-hour free trial · 20 credits · payment method binding required
Free for first 24 hours · then billed normally
Personal
More power and credits for daily productivity.
Pro
Maximum compute for heavy workloads and teams.
Model Pricing
All models, transparent credit cost
Every plan includes all models below. Choose the right balance of speed and capability for each task.
| Model | Provider | Input | Output | Cached | Output × | Cache × |
|---|---|---|---|---|---|---|
GLM-4.5 Flash Fastest, ideal for simple tasks | Zhipu | 0.000 | 0.000 | — | — | — |
Gemini 3.1 Flash-Lite Ultra-fast, lightweight tasks | Google20% OFF | 1.150.92 | 6.755.40 | 0.120.09 | 5.87× | 0.1× |
GPT-5.4 Nano Compact, quick responses | OpenAI20% OFF | 1.000.80 | 6.255.00 | 0.100.08 | 6.25× | 0.1× |
GLM-4.6 Balanced Chinese LLM | Zhipu | 3.00 | 11.01 | 0.54 | 3.67× | 0.18× |
Gemini 3.1 Flash Balanced speed & quality | Google20% OFF | 1.251.00 | 7.506.00 | 0.500.40 | 6× | 0.4× |
GPT-5.4 Mini Efficient, daily use | OpenAI20% OFF | 3.753.00 | 22.5018.00 | 0.380.30 | 6× | 0.1× |
GPT-4.1 Strong reasoning, complex tasks | OpenAI20% OFF | 10.008.00 | 40.0032.00 | 2.502.00 | 4× | 0.25× |
Gemini 3.1 Pro Preview Latest Gemini capabilities | Google20% OFF | 18.0014.40 | 81.0064.80 | 1.801.44 | 4.5× | 0.1× |
GPT-5.2 Turbo Most capable OpenAI model | OpenAI20% OFF | 8.757.00 | 70.0056.00 | 0.880.70 | 8× | 0.1× |
GPT-5.4 Next-gen OpenAI flagship | OpenAI20% OFF | 12.5010.00 | 75.0060.00 | 1.251.00 | 6× | 0.1× |
Claude Sonnet 4.6 Excellent coding & analysis | Anthropic | 16.50 | 82.50 | 1.65 | 5× | 0.1× |
Claude Opus 4.6 Top-tier reasoning & creativity | Anthropic | 25.00 | 125.00 | 2.50 | 5× | 0.1× |
Monthly Credits Reset
Credits renew at the start of each billing cycle. Add-on packs never expire.
No Lock-in
Upgrade, downgrade, or cancel anytime. Your data is always exportable.
Add-On Credits
Running low? Top up instantly
Sign in and continue in Wallet to purchase one-time credit packs.
100credits total
$0.10 per credit
A small top-up for occasional extra tasks.
300credits total
$0.10 per credit
Great value for a full month of heavy automation.
900credits total
700 base + 200 bonus
$0.10 per credit
900 total credits. 700 base + 200 bonus — best for heavy workloads.
Compare Plans
Everything side by side
| Feature | Starter | PopularPersonal | Pro |
|---|---|---|---|
| UsageCredits / month | 100 | 200 | 800 |
| Max task duration | 30 min | 1 hr | Unlimited |
| Concurrent tasks | 2 | 5 | Unlimited |
| ComputevCPU cores | 2 cores | 4 cores | 8 cores |
| RAM | 4 GB | 8 GB | 16 GB |
| SSD storage | 40 GB | 80 GB | 160 GB |
| ModelsAll AI models | |||
| Gemini Flash-Lite | |||
| Gemini Flash | |||
| GPT-4.1 | |||
| Gemini Pro | |||
| GPT-5.2 | |||
| GPT-5.4 | |||
| Claude Sonnet 4.6 | |||
| Claude Opus 4.6 | |||
| AccessWeb dashboard | |||
| Telegram bot | |||
| Slack integration | Not included | Included | Included |
| FeaturesConversation history | 30 days | 90 days | Unlimited |
| Usage analytics | |||
| SupportPriority support | |||
| Dedicated onboarding |
Trusted by users
What our users say
“The Personal plan pays for itself in saved time every week. My agent handles competitor monitoring and weekly summaries automatically.”
James R.
Solo Founder
“I love that all models are available on every plan. I use Gemini Flash-Lite for quick drafts and GPT-5.2 when I need the best output.”
Aiko M.
Content Strategist
“Having dedicated compute makes a real difference. My agent processes tasks fast and I never worry about shared resource contention.”
David K.
Engineering Lead
FAQ
Pricing questions
What counts as one credit?
Credits are consumed based on the AI model used and the number of tokens processed. 1 credit covers approximately 33,000 tokens (input + output combined). Different models have different multipliers: GLM-4.5 Flash is free (0×), Gemini Flash-Lite and GPT-5.4 Nano cost 1 credit per 1M tokens (1×), while Claude Opus 4.6 costs 21 credits per 1M tokens (21×). Check the model pricing table above for the full list.
Do all plans get access to all AI models?
Yes — every paid plan includes access to all listed AI models. The difference between plans is the compute resources and monthly credit allocation.
What does the dedicated compute mean?
Each plan includes reserved server resources (vCPU, RAM, SSD) that run your AI agent 24/7. Starter gets 2 vCPU / 4 GB RAM, Personal gets 4 vCPU / 8 GB RAM, and Pro gets 8 vCPU / 16 GB RAM — all with dedicated SSD storage.
Can I upgrade or downgrade anytime?
Yes. You can switch plans at any time from your Settings page. Upgrades take effect immediately; downgrades apply at the start of your next billing cycle.
What happens if I run out of credits mid-month?
Your agent pauses gracefully and you receive a notification. No additional charges are incurred. You can purchase a top-up pack instantly or wait for your credits to reset at the start of the next billing period.
Do unused credits roll over?
Monthly plan credits reset each billing cycle and do not roll over. However, add-on credit packs never expire and stack on top of your monthly allocation.
Can I connect my own API keys?
Not currently. KeepClaw manages all model integrations so you never have to handle API keys, rate limits, or billing with individual providers. This is by design — one bill, zero ops.
Still have questions? Email us
Launch your agent in 60 seconds
Dedicated compute, all AI models, zero ops. Pick a plan and your agent is live instantly.