unbiased/paretoClosed APIBlended multimodal model that runs multiple language models in parallel and returns a synthesized answer
Benchmark Pricing$2.50 / 1M in$7.50 / 1M out$0.25 cache read
Context: 262,144 tokens•Max Output: 131,072 tokens•Released: 2026-09-17
TECHNICAL SPECIFICATIONS
Context Window262,144262.1K tokens
Max Output131,072131.1K tokens
LicenseProprietaryCloud API only
Input Modalities
TextVision
Function / Tool CallingSupported (MCP Ready)
Structured OutputsGuaranteed Schema
MODEL CONTEXT PROTOCOL (MCP) READINESS
✓ Pareto natively supports OpenAI / Anthropic format tool calling and is fully compatible with MCP servers and agent loops.
MCP Context Window Recommendations
With a context capacity of 262,144 tokens, this model can handle moderately large MCP tool definitions and multi-turn conversation logs comfortably.
API CALL INVOCATION EXAMPLE
// Example: Calling Pareto via OpenAI SDK
import OpenAI from "openai";
const client = new OpenAI({
apiKey: process.env.UNBIASED_API_KEY,
});
const response = await client.chat.completions.create({
model: "unbiased/pareto",
messages: [{ role: "user", content: "Hello! Can you inspect MCP tools?" }],
});
console.log(response.choices[0].message.content);PRICING BY PROVIDER
NanoGPTunbiased/pareto-26.9
$2.50 / $7.50$0.25 cache
AI/ML APIunbiased/pareto
$3.44 / $10.32$0.34 cache
Cloudflare AI Gatewayunbiased/pareto
$2.50 / $7.50$0.25 cache
Kilo Gateway~unbiased/pareto-latest
$0.80 / $3.20$0.03 cache
OpenRouter~unbiased/pareto-latest
$0.80 / $3.20$0.03 cache
Rates in USD per 1M tokens. Rates synchronized from models.dev public catalog.
Prompt Caching Active
This model supports prompt caching at $0.25 / 1M tokens, saving up to 90% on repetitive MCP tool definitions and system instructions.