All models

Unbiased

Pareto 26.10 balanced

unbiased/pareto-26-10

Pareto is a multimodal composite model built for research, coding, and agentic workflows, while delivering frontier-level performance across a broad range of general-purpose tasks.

Input / 1M

$0.80

USD per million tokens

Output / 1M

$3.20

USD per million tokens

Est. cost / token

$0.0000014

$1.40 per 1M, 3:1 mix

Context window

1.05M

max output 131K

Major features

VisionCachingTools

Task suitability (heuristic 1–5, not a benchmark)

  • coding4
  • reasoning3
  • math3
  • summarization4
  • extraction4
  • classification4
  • chat4
  • creative4
  • translation4
  • vision4
  • long context5
  • agentic4

Price history

RecordedInputOutputCachedSource
2026-11-05 03:20$0.80$3.20$0.030openrouter

A row is added whenever a sync detects a price change. Vendor pricing source

Pricing (USD per 1M tokens)

Input
$0.80
Output
$3.20
Cached input
$0.030
Blended (3:1)
$1.40

Details

Company
Unbiased
Version
26.10
Family
Pareto
Released
2026-10-15
Parameters
Undisclosed
Max output
131K
Open weights
No
Inputs
image, text
Last synced
35 min ago