r/GeminiFeedback

Tested Gemini 3.1 Flash Live for production voice calls, the feel is noticeably better but latency claims need context
▲ 3 r/GeminiAI+1 crossposts

Tested Gemini 3.1 Flash Live for production voice calls, the feel is noticeably better but latency claims need context

Been building voice agents for a while now, and integrated Gemini 3.1 Flash Live into our open source stack as soon as the API went live. Wanted to share some honest observations

The good stuff first. The voice cadence and overall feel of calls is genuinely better than what you get from the classic STT + LLM + TTS pipeline. Turn-taking feels more natural. Interruptions are handled way more gracefully. The model just "gets" conversational rhythm in a way that stitching together STT + LLM + TTS never really achieved. Cost also looks very competitive, which matters a lot in S2S.

Now the stuff nobody seems to be talking about. We averaged around 922ms latency end-to-end in our testing. That's not bad, but it's not the sub-300ms numbers I've seen some people throw around. We were testing from Asia, so region probably plays a role here. Would love to know what others are seeing from US/EU

The other thing that caught us off guard is transcripts. You can't access them live during the call, only after it's done. If you're doing any kind of context stitching or real-time context engineering during conversations, this makes things harder. 

Honestly though I don't think we're going back to the old pipeline. The quality gap in how the conversation actually feels is too big.

We integrated this into Dograh, our open-source voice agent platform (very much like Vapi) , if anyone wants to try it themselves: https://github.com/dograh-hq/dograh

What latency numbers are others getting? And has anyone found a clean workaround for the live transcript limitation?

u/Slight_Republic_4242 — 3 hours ago
▲ 5 r/GeminiCLI+2 crossposts

Gemini CLI hangs silently on startup and never respond, process is alive but frozen

I’ve been using Gemini cli working on my Linux machine (Ubuntu , node v22.22.1) and I am stuck. I have 0.37.1 installed and the problem is that it hangs indefinitely on any command. The process is slaws and stay alive , but it never actually does anything - it just freeze silently. The version check and help command work fine, and I can reach google’s API without issues , but the moment I try to run interactive mode or use the headless -p flag , it just hangs forever with no error message at all.

I’ve tried everything. Full uninstall and reinstall, deleted and recreated my config from scratch , even updated from 0.36.0 to 0.37.1. Thinking it was a version bug. I created file permisssion, clean up temp directories , verify network connectivity- nothing helps. It just sit there waiting for that never comes. The 0Auth flow also hangs during authentication, which tells me it’s not the auth problem but something deeper in the initialization.

I’m just sharing my experience here in case anyone else has hit this. Thanks to everyone who might have suggestions or insights. If you faced the same issue with Gemini Cli hanging like this ,

did you figure out what was causing it ?

What worked for you ?

Any help would be greatly appreciated.

u/KoTaw_Ray — 8 hours ago
▲ 6 r/GeminiAI+1 crossposts

What the hell happened to the Gemini Model Quota on the Google AI Pro Plan?

Working on a project in Antigravity. 5 prompts in total. Entire quota exhausted, used 300+ AI Credits and now my gemini pro 2 usage refreshes in 5 days.

I've been using the Gemeni AI pro plan for 3+ months now and never experiences a usage drain like this.

Oh, and the nice thing is that Gemini 3 pro is so dumb now, that it made more errors than fixed.

Is this Google's new business model? Tard out the model, doesn't build or fix anything, repeat, profit?

CAN'T CANCEL THIS GARBAGE PLAN FAST ENOUGH!

reddit.com
u/nobodyhere3369 — 10 hours ago

Image compression on uploaded images has gotten worse. Screen reading is broken.

I’ve relied heavily on Gemini for help with complex UIs and form-filling by uploading full-page screenshots (taken using the Firefox screenshot tool). It used to be a lifesaver, but lately, the image compression seems incredibly aggressive.

The text is becoming so blurred that the AI can no longer read the screen. Instead of admitting it can’t see the details, it has started hallucinating answers based on what it expects to see on a standard page. It’s essentially broken one of my primary use cases.

Is anyone else having the same issue?

reddit.com
u/stats1101 — 3 hours ago

Image Generation Went To Crap

Paid Pro User here. Image generation looks like dog excrement. Just in the last two days, it's totally gone downhill. Less detail is visible in the Pro pictures now.

reddit.com
u/Common-Ruin-7146 — 7 hours ago
▲ 9 r/GeminiAI+1 crossposts

Paid Gemini Pro is not available

I’m a bit frustrated and wanted to check if others are experiencing the same thing.

I’m paying for Gemini Pro mainly to use the Pro model, but today I wasn’t able to access it at all. It kept falling back to the free version, and I got a message saying Pro is currently unavailable due to high demand.

I understand services can get busy, but if I’m paying specifically for Pro access and can’t use it when I need it, it does make me question the value of the subscription. The extra perks like storage or other Google benefits aren’t really relevant to me. I subscribed purely for the AI capability.

If this continues, I’ll probably end up cancelling.

Is anyone else running into this issue, or is it just me?

reddit.com
u/B4Beta — 23 hours ago
Week