Flex and Priority processing for Gemini models | OpenRouter | Page 1

mystic stump Apr 3, 2026, 2:34 AM

#

Pls
https://x.com/i/status/2039782721460027676

Balance cost & reliability with our new Flex & Priority inference tiers in the Gemini API!

Flex: Pay 50% less for cost-sensitive & latency-tolerant workloads
Priority: Highest reliability for your most critical, interactive apps (with premium pricing)

Together with the async

▶ Play video

mystic stump Apr 4, 2026, 3:23 AM

#

bump

#

is this implemented?

lament gale Apr 4, 2026, 4:12 PM

#

Bump, and also, this is a common feature on other providers too like openAI which has basically the same feature and pricing scheme. I would like it on every major provider that support supports it. I believe Azure and Amazon Bedrock also have some models that support this. Let me use a :flex slug

wraith pike Apr 5, 2026, 3:25 AM

#

in support website it says it supports openai but not google....tested and didn't see changes or pass through for the flex para

mystic stump Apr 7, 2026, 10:56 PM

#

bump

#

we just want transparency man how hard is it to support flex processing for Google models, flash is costing me a ton

mystic stump Apr 8, 2026, 8:42 AM

#

bump

indigo flame Apr 9, 2026, 8:33 AM

#

openrouter finished. I found a solution and already setup LiteLLM and OpenWebUI and its working.

mystic stump Apr 10, 2026, 1:24 AM

#

does it work?

#

I'm mainly using G3 flash for chat

#

which.... in terms of cost its not appealing

mystic stump Apr 11, 2026, 4:25 AM

#

any updates?

#

bump

indigo flame Apr 11, 2026, 4:30 PM

#

I'm spending SIGNIFICANTLY less with my setup. a few weeks ago I realized i'm paying2-3x times more than other developers using the same models. Now that I switched over OMG what a cost difference without any performance difference

#

the set up took a couple hours but the product is significantly better. Sure, I have to update myself and add new models as they come, but at least i wont experience the significant downtime and expense of using this crap platform.

#

FYI, i loved OpenRouter and recommended it to several friends. but there are limits and shutting out users and never updating us on whats going on is unacceptable

wraith pike Apr 11, 2026, 11:29 PM

#

BaliHoo, get lost will you?

subtle finch Apr 16, 2026, 10:45 AM

#

bump

mystic stump Apr 16, 2026, 12:08 PM

#

bumpy bump bump meowslightsmile

#

meowslightsmile

#

meowslightsmile

indigo flame Apr 17, 2026, 3:30 AM

#

You folks need to jump this sinking ship. You’re paying wayy too much for subpar service and no guarantee of uptime

mystic stump Apr 17, 2026, 4:26 AM

#

using official google sdks is pita to work with though doable

wraith pike Apr 19, 2026, 6:36 AM

#

where is the confirmation from the team? is this like soon, never or bottom of the priority list?

keen sphinx Apr 22, 2026, 1:12 AM

#

bump

mystic stump Apr 26, 2026, 4:03 AM

#

bump

mystic stump Apr 26, 2026, 4:03 AM

#

wraith pike where is the confirmation from the team? is this like soon, never or bottom of t...

ngl this shouldn't be hard to implement or clarify at all

#Flex and Priority processing for Gemini models