Jylo AI Model Reference Card Jylo AI Model Reference Card

Jylo AI Model Reference Card

Aaron Kirk Aaron Kirk

Jylo hosts AI models from several providers. Which you can see depends on your region and on what your administrator has enabled.

How to Choose

Capability is described in tiers: routine for straightforward, high-volume work; professional for most analysis; complex where depth matters more than speed or cost. Each model's description in the interface names its tier.

Cost is a three-point scale, relative rather than absolute — it shows where a model sits against the others, not a price.

Context is how much text a model can process in a single interaction.

Models Available

OpenAI, hosted on Microsoft Azure

Model

Description

Tool use

Cost

Context

GPT-4.1

Capable all-rounder for routine tasks

No

$

1M

GPT-5.4

Skilled daily driver for professional tasks

Yes

$$

1M

GPT-5.5

Master polymath for complex tasks

Yes

$$$

1M

GPT-5.6 Luna

Proficient daily driver for professional tasks

Yes

$

1M

GPT-5.6 Terra

Versatile all-rounder for professional tasks

Yes

$$

1M

GPT-5.6 Sol

Formidable virtuoso for complex tasks

Yes

$$$

1M

o3-mini

Concise specialist for STEM fields

No

$

200K

o3

Rigorous specialist for STEM fields

No

$$

200K

Anthropic, hosted on Amazon Web Services

Model

Description

Tool use

Cost

Context

Claude Haiku 4.5

Swift generalist for routine tasks

No

$

200K

Claude Sonnet 4.6

Adept all-rounder for professional tasks

Yes

$$

1M

Claude Opus 4.6

Consummate authority for complex tasks

Yes

$$$

1M

Web search runs only on Azure-hosted models, so it is unavailable on the Claude models even though Sonnet 4.6 and Opus 4.6 are otherwise tool-capable. The o-series cannot call tools at all.

Our Recommendations

Add comment

Please sign in to leave a comment.