Start free. Pay for more review.
Hobby includes two Refinement reviews each month. Paid plans include compute credits for model usage, more review formats, and higher limits. The models and rounds you choose determine how far your credits go.
Hobby
- Monthly allowance2 Refinement reviews / month
- Rounds per review2
- Model catalogFree-tier models
- Review formatsRefinement
- Document context2,000 tokens
- Thinking + web searchNot included
- API accessNot included
Pro
- Monthly allowance1,000 compute credits / month
- Rounds per review5
- Model catalogAll 100+ models
- Review formatsAll 7 formats
- Document context1 million tokens
- Thinking + web searchIncluded
- API accessREST API
Pro Max
- Monthly allowance5,000 compute credits / month
- Rounds per review10
- Model catalogAll 100+ models
- Review formatsAll 7 formats
- Document context5 million tokens
- Thinking + web searchIncluded
- API accessREST API + webhooks
| Feature | Hobby $0/mo | Pro $20/mo | Pro Max $100/mo |
|---|---|---|---|
| Monthly allowance | 2 Refinement reviews / month | 1,000 compute credits / month | 5,000 compute credits / month |
| Rounds per review | 2 | 5 | 10 |
| Model catalog | Free-tier models | All 100+ models | All 100+ models |
| Review formats | Refinement | All 7 formats | All 7 formats |
| Document context | 2,000 tokens | 1 million tokens | 5 million tokens |
| Thinking + web search | Not included | Included | Included |
| API access | Not included | REST API | REST API + webhooks |
| Start free |
Hobby tier is permanent. We will never retroactively bill your account or revoke historical runs.
Any amount. Any time.
No subscription? No problem. A positive balance unlocks all models and all 7 debate formats in the web app. Credits are applied instantly and never expire. Subscription users can also buy extra credits when their monthly allowance runs out.
You'll receive 1000 credits · Min $5 · Max $10,000
Questions, answered.
What is included in the free plan?
Two Refinement reviews each month, using free-tier models and up to two rounds per review. You can also browse public reviews and panel templates. Choose a paid plan or add prepaid credits when you need more reviews.
What is a compute credit?
One compute credit represents $0.01 of metered model-provider cost. A run uses more credits when it uses more models, rounds, tokens, web search, or expensive models. API usage is metered at twice the provider cost. The audit trail shows the actual usage for every run.
How many reviews will my credits cover?
There is no fixed number. Usage depends on the models, input length, response length, rounds, and optional tools. Check the estimate before starting and the recorded usage afterward; an estimate can differ from the final cost.
Can I use RigorDesk from Claude, ChatGPT, or Codex?
Yes. RigorDesk exposes its review workflows through the Model Context Protocol (MCP) at https://api.rigordesk.com/mcp. Connect it as a custom connector in Claude, a custom app in ChatGPT (Business, Enterprise, or Edu with developer mode), or a remote MCP server in Codex, then sign in with your RigorDesk account. Setup takes a few minutes and is covered in the docs under Connect with MCP.
How does MCP usage bill?
Exactly like the web app and REST API. Creating panels, listing models, and reading results is free. Each review started through MCP consumes your subscription allowance first, then prepaid credits. Check your balance anytime with the get_usage tool before starting a run.
Can I switch tiers mid-month?
Yes. Upgrades take effect immediately and the remaining value on your current plan is credited to the new one. Downgrades take effect at the next billing cycle.
Do you offer team plans or custom contracts?
Current subscriptions are for one account. For team access, consolidated billing, or a custom contract, contact us at sharjeel@tellodb.com.
What are prepaid credits and how do they work?
Prepaid credits are one-time purchases that let you run reviews without a subscription, or extend your usage after your monthly subscription allowance runs out. Credits never expire. Usage is deducted from your subscription first, then from credits.
Can I use prepaid credits to access higher-tier models?
Yes. A positive prepaid credit balance unlocks all models and all 7 review formats in the web app. Round depth, thinking, web search, larger documents, API access, and webhooks still follow your subscription tier.
What do DeepSeek V4 Pro 0813 and Grok 4.6 cost?
DeepSeek V4 Pro 0813 lists at $0.435 per million input tokens and $0.87 per million output tokens. Grok 4.6 lists at $2 per million input tokens and $6 per million output tokens; above 200,000 prompt tokens, its provider rates increase to $4 and $12 respectively. Pre-run estimates use the standard rate and can understate those long Grok prompts; the audit trail meters the actual provider-reported cost.
What does Gemini 3.8 Flash cost?
Gemini 3.8 Flash is available on Hobby, Pro, and Pro Max. Estimates use its standard provider rate of $1.50 per million input tokens and $7.50 per million output tokens, excluding the temporary 50% OpenRouter promotion. API usage is metered at twice provider cost; completed runs record actual provider-reported usage.
What do GPT-6.1 Sol, GPT-6 Luna, and Claude 5.5 cost?
Provider input/output rates per million tokens are $2/$10 for GPT-6.1 Sol and Claude Sonnet 5.5, $0.10/$0.50 for GPT-6 Luna, and $4/$20 for Claude Opus 5.5. Luna is available on Hobby; the other three require Pro, Pro Max, or prepaid credits. Sol and Luna have 1,050,000-token context windows; both Claude models have 1,000,000. Sol and Luna rates increase above 272,000 prompt tokens.
What do DeepSeek V4.1 Flash and MiMo V2.6 Pro cost?
Provider input/output rates per million tokens are $0.0198/$0.396 for DeepSeek V4.1 Flash and $0.435/$0.87 for Xiaomi's MiMo V2.6 Pro. Both have 1,048,576-token provider context limits. DeepSeek is available on Hobby; MiMo requires Pro, Pro Max, or prepaid credits.
What do Grok 4.7, Muse Spark 1.3, and Command A+ cost?
Provider input/output rates per million tokens are $2/$6 for Grok 4.7, $1.25/$4.25 for Muse Spark 1.3, and $0.30/$1.50 for Command A+. All three require Pro, Pro Max, or prepaid credits. Grok rates increase to $4/$12 above 200,000 prompt tokens. Provider rates exclude RigorDesk's usage multiplier and additional tool charges.
What does GPT-6 Astra cost?
GPT-6 Astra lists at $5 per million input tokens and $25 per million output tokens with a 1,050,000-token context window. It is available to Pro and Pro Max subscribers as a frontier model, or via prepaid compute credits.