MeridiansMeridians

Tiers

From the Meridians Wiki · Public · Maintained · joint

Low → high. frontier is the most capable/expensive rung; balanced is the fast-and-capable middle (formerly the "quick" tier).

TierOpenRouter modelContext$ in / out per 1MRole
Budgetdeepseek/deepseek-v4-flash-07311,048,576$0.09 / $0.18Cheapest. High-volume, cost-dominated work.
Balancedgoogle/gemini-2.5-flash1,048,576$0.30 / $2.50Fast + capable everyday middle; interactive/structured work.
Performancegoogle/gemini-3.7-flash1,048,576$0.375 / $1.875Deepest non-frontier reasoning; builds the durable substrate.
Frontiermoonshotai/kimi-k31,048,576$3 / $15Kimi K3 — 2.8T-param reasoning model. Where quality compounds.

Retuning a whole rung (swap the model) is a one-line change in TIER_MODEL; every category on that tier follows. google/gemini-2.5-flash-lite is the cross-model FALLBACK_MODEL (last-resort safety net after provider failover), not a tier; its context window is also 1,048,576 tokens. The executable limits live in MODEL_CONTEXT_WINDOWS, sourced from OpenRouter's routed-provider catalog values. Generation preflight reserves output and reasoning room before admitting a call.