generated: '2026-08-30' method: searched source: https://www.tellius.com/pricing name: Tellius plans and pricing currency: null currency_note: 'No currency is stated because no price is stated. Both tiers show "Customized".' billing_model: 'Capacity-based, not per-user. The marketing llms.txt states "Capacity-based pricing with no per-user fees. Contact for details." Volume discounts and term commitments are available.' public_prices: false plan_count: 2 plans: - name: Premium price: null price_label: Customized currency: null billing_period: null users: 'Up to 10 Users' limits: live_pushdown_query: 'Up to 50M rows in live-mode' in_memory_cache: '10GB data storage' automl: '5 Concurrent Jobs' data_pipelines: '5 Concurrent Jobs' feeds: 'Limited' overage: null overage_note: 'No overage pricing is published.' - name: Enterprise price: null price_label: Customized currency: null billing_period: null users: 'Unlimited Users' limits: live_pushdown_query: 'Unlimited' in_memory_cache: 'Unlimited' automl: 'Unlimited' data_pipelines: 'Unlimited' feeds: 'Unlimited' overage: null trial: available: true length: '30 days' url: https://www.tellius.com/free-trial note: 'Marketing llms.txt: "30-day free trial. Production-ready in 4-6 weeks."' api_pricing: metered: false note: 'Neither the REST API nor the MCP server is separately priced or metered. API and MCP access ride the platform entitlement, and the capacity dimensions that bound them are the tier limits above (concurrent jobs, in-memory cache, live-mode row ceiling) rather than a request quota.' findings: - 'Prices are contact-sales only, but the QUOTAS are published per tier, which is more than most enterprise analytics vendors disclose. Those quotas are the real limits an API integrator hits: 5 concurrent AutoML jobs on Premium bounds how many ML prediction pipelines can run at once, and the 50M-row live-mode ceiling bounds what a tellius_ask question can scan.' - 'No self-serve purchase path and no published per-seat or per-call rate.'