Reconciled LLM Pricing. For Agents & Humans.

Multi-source verified token pricing with full history, real-time change feed, and agent-native endpoints. Replace manual spreadsheets with a single trusted API.

Browse Models View Features

2,430models tracked

121providers

<60s agolast price change

OpenAIgpt-4oin $2.50/M · out $10.00/M

Anthropicclaude-3-5-sonnetin $3.00/M · out $15.00/M

Googlegemini-1.5-proin $1.25/M · out $5.00/M

OpenAIgpt-4o-miniin $0.15/M · out $0.60/M

Anthropicclaude-3-haikuin $0.25/M · out $1.25/M

Googlegemini-1.5-flashin $0.075/M · out $0.30/M

Mistralmistral-largein $2.00/M · out $6.00/M

Coherecommand-r+in $2.50/M · out $10.00/M

DeepSeekdeepseek-r1in $0.55/M · out $2.19/M

Metallama-3.3-70bin $0.20/M · out $0.60/M

OpenAIgpt-4oin $2.50/M · out $10.00/M

Anthropicclaude-3-5-sonnetin $3.00/M · out $15.00/M

Googlegemini-1.5-proin $1.25/M · out $5.00/M

OpenAIgpt-4o-miniin $0.15/M · out $0.60/M

Anthropicclaude-3-haikuin $0.25/M · out $1.25/M

Googlegemini-1.5-flashin $0.075/M · out $0.30/M

Mistralmistral-largein $2.00/M · out $6.00/M

Coherecommand-r+in $2.50/M · out $10.00/M

DeepSeekdeepseek-r1in $0.55/M · out $2.19/M

Metallama-3.3-70bin $0.20/M · out $0.60/M

OpenAI

Anthropic

Google

OpenRouter

LiteLLM

Hugging Face

Full history. Multiple sources. Zero guesswork.

Most pricing tools give you a snapshot. We give you the full timeline with source attribution.

Full Price History

Every model, every change — timestamped and immutable. Not just a snapshot, the complete timeline.

Multi-Source Reconciliation

OpenRouter + LiteLLM + Hugging Face. 2-source agreement required before publishing. Flagged discrepancies logged.

Agent-Optimized APIs

/v1/context for clean system prompts, /v1/ask for NL queries, SSE stream for live deltas.

Real-Time Change Feed

Price changes surfaced within 60 seconds. Trust metadata — confidence, source, age — on every record.

Built for LLM agents

Dedicated endpoints designed to fit in agent system prompts, handle natural language queries, and stream price deltas in real time — without blowing your token budget.

Explore Features

GET/v1/context

~2k token pricing snapshot for agent system prompts

POST/v1/ask

Natural language → structured pricing response

GET/v1/stream/changes

SSE stream with Last-Event-ID reconnection

GET/v1/recommend

Ranked model recommendations by task, context, and price

Feature highlights for agents and developers.

Everything is available through one unified API surface.

/v1/contextAgent-optimized

Context injection for system prompts

Your agent's system prompt stays lean. One endpoint returns a ~2k token pricing snapshot — structured, current, and small enough to inject without blowing your context window.

Toggle to compare with vs. without

// system prompt
You are a cost-optimization agent.

{{pricing_context}}  ← injected from /v1/context

// resolves to:
LLM_PRICES (as of 2026-03-14T09:41Z)
gpt-4o          $2.50/1M in  $10.00/1M out  conf:high
claude-sonnet-4  $3.00/1M in  $15.00/1M out  conf:high
gemini-2.0-pro  $1.25/1M in   $5.00/1M out  conf:med
llama-4-maverick $0.20/1M in   $0.60/1M out  conf:high
... 118 more models

~2,100 tokens120+ models covered

/v1/stream/changesReal-time

Live price change stream

SSE endpoint with replay semantics. Every price delta surfaces within 60 seconds — with model, old/new price, source, and confidence metadata. Disconnect and reconnect without missing events.

Try disconnecting and reconnecting

GET/v1/stream/changes

connected

/v1/askNatural language

Ask pricing questions in plain English

No query syntax to learn. Post a natural language question, get back structured JSON with model recommendations, pricing breakdowns, and source attribution. Built for agent tool-use flows.

Click an example or type your own

POST/v1/ask

// response will appear here

Start tracking LLM prices today

Start exploring models immediately. Use the full API surface for history, recommendations, stream updates, and webhook automation.

Browse Models Explore Features