Skip to content
OperatorNest

AI model picker

Choose a model class for a task, then verify current provider facts.

Example: Sam, founder: Research with web sources selected; starting recommendation is a frontier general model with search.

Describe the task

Constraints

Starting recommendation

Frontier general with web search

Research needs source finding, cross-checking and a clear synthesis. Use web search as a separate capability, then ask the model to show which source supports each claim.

Example: Sam is checking current market claims, so this starting point pairs a frontier general model with web search.

Constraint: best quality for this task.

Example models

Provider-stated details, linked field by field.

Checked 28 September 2026

Google · Reasoning

Nano Banana Pro

Checked 28 September 2026

Context
65,536 input tokens Source
Inputs
text, image Source
Relative price tier
$2 input and $120 output per 1M tokens Source
Open weights
No Source

Anthropic · Reasoning

Claude Sonnet 5

Checked 28 September 2026

Context
1,000,000 input tokens Source
Inputs
text, image, pdf Source
Relative price tier
$2 input and $10 output per 1M tokens Source
Open weights
No Source

Model facts from models.dev (MIT).

How it works

Task type sets an initial model class. Constraints can favor a smaller model, stronger reasoning, a long input window or downloadable weights.

The picker uses no benchmark scores and does not claim that one provider is best.

Sources and checked dates

Limits

  • Recommendations are starting points, not quality rankings.
  • Context limits are provider-stated input limits, not a promise every detail is used correctly.

Common questions

Does the picker name one best model?
No. It recommends a model class based on task shape and your constraints. Compare the linked provider facts, then test the choice on your own examples.
Does a long context window mean a model will read everything well?
No. It means the provider lists room for that much input. Relevant detail can still be missed, so check summaries against the source.
Does web research come built into every model?
No. A web search or browsing tool supplies current sources. This picker treats that as separate from the model itself.

OperatorNest can take on repeat work, check with you before consequential steps, and leave a receipt for the result. See how an always-on operator works.

Hand off your first task tonight.

Tell us your email and what you'd hand off first. We'll send your access details and help you set up your operator.