// Interactive Tool

Model Routing
Calculator

Right model. Right cost. No guesswork.

Not every task needs the most expensive model. Configure your workload parameters below and get a routing recommendation with real cost estimates across four model tiers.

// Section 01

Workload Parameters

Quick presets

Task Type

Complexity

Calls per day

Avg input tokens

Avg output tokens

Latency Requirement

Quality Floor

// Section 02

Recommendation

Sonnet / GPT-4o

Claude Sonnet 4, GPT-4o, Gemini 2.5 Pro

$13.50

per month

Routing Strategy

Start with Tier 2 (Sonnet / GPT-4o) for all requests. Escalate to Tier 3 (Opus) for failures, low-confidence outputs, or complex edge cases.

Daily calls

100

Cost per call

<$0.01

Latency

1-5s

Monthly calls

3,000

Cost Comparison — All Tiers

Same workload (100 calls/day, 500 in / 200 out tokens)

Tier 1

Flash / Haiku

$1.13/mo

Tier 2Pick

Sonnet / GPT-4o

$13.50/mo

Tier 3

Opus

$67.50/mo

Tier 4

Reasoning (o1/o3)

$58.50/mo

Per 1M tokens pricing

Tier	Models	Input	Output	Monthly Est.
T1	Gemini 2.0 Flash, Claude Haiku 3.5	$0.25	$1.25	$1.13
T2	Claude Sonnet 4, GPT-4o, Gemini 2.5 Pro	$3	$15	$13.50
T3	Claude Opus 4	$15	$75	$67.50
T4	o1, o3, Claude with extended thinking	$15	$60	$58.50