Based on several use cases, Gemini-3-Pro-preview is demonstrating substantially inferior performance compared to Gemini 2.5 Pro specifically when processing long context content interactions - particularly large file uploaded content. The degradation is likely attributable to it’s inability to per…

[image] JSON_B_Kidd: he degradation is likely attributable to it’s inability to perform at low temperature settings - this is likely the culprit affecting the model’s ability to think critically and sharply and maintain coherence and accuracy across long context windows. Hi @JSON_B_Kidd , We…

Subject: Re: Real-world data on iPhone 15 Pro Max confirms thermal throttling issues with long-context usage Hi there, I strongly agree with your hypothesis regarding temperature affecting performance. I just experienced a concrete example of this while stress-testing Gemini on mobile. Device Spe…

Couldn’t agree more. I do a lot of general research then I end up creating a summary report. Then I do a ton of iterations because I ask more questions, reorganize content, manage change delete content. Google Gemini is dropping content, getting off track, changing content (with no authorization) …

By temperature the original poster probably didn’t mean device temperature, but rather model temperature, which is an input parameters to large language models and control the creative freedom the model could use to answer questions.

I’m not a developer but Gemini 3 is telling me it has a fatal flaw and can’t be trusted. Kinda freaking me out. Lol

I 100% agree here. The first week or two, Gemini 3.0 seemed leaps and bounds above 2.5. It was able to maintain long conversations, talk intelligently and recall items from much early in the conversation and I felt was a huge upgrade over 2.5. Unfortunately over the last week or so I feel like the…

Were you able to find anything reliable? I’ve started to hit a problem this week where what used to be routine files well within the context window are now not being fully “grabbed”, with notices that my file size is too large and the code is being truncated, preventing proper review. Very annoying…

No, unfortunately, it seems like it’s somewhat ‘by design’. [image] New Model Levels (Fast/Thinking/Pro) Continue to Be a Problem for Long Term Projects Google AI Studio Today I noticed there’s now three ‘model’ options available to choose from, Fast, Thinking…

Hmmmm. Thanks. At least it’s not just me lol That’s a big frustration. It’s definitely advertised to be able to handle things like 50,000+ lines of code, and as recently as last week seemed to be doing remarkable. This week, I’ve had serious issues with it truncating files I’ve uploaded and not sca…

Hello, even after pro came out, this bug still persists. It has made gemini as a product unusable for my specific daily needs. That’s unfortunate as before the dec 4 update, gemini 3 was perfect. I really hope that htis’ll be fixed soon.

Gemini 3 significantly worse thant 2.5 Pro at long context. Temperature likely to blame

Google AI Studio

MBB December 17, 2025, 3:40pm 16

No, unfortunately, it seems like it’s somewhat ‘by design’.

Topic		Replies	Views
Gemini 3 review after 1 month: inconsistent at best, poor at worst compared to Gemini 2.5 Google AI Studio gemini-3	9	3067	February 16, 2026
Feedback: Gemini 3 Pro Preview - Significant regression in Reasoning, Context Retention, and Safety False Positives compared to 2.5 Google AI Studio feedback , models	4	454	May 8, 2026
New Model Levels (Fast/Thinking/Pro) Continue to Be a Problem for Long Term Projects Google AI Studio ai-studio , feedback , gemini	11	4596	April 4, 2026
Gemini 2.5 Pro's Response Quality Regression Google AI Studio models , gemini-25 , gemini-2-5	9	1401	September 15, 2025
Gemini 3.0 Pro is ignoring my current prompts and repeating old answers in longer chats Google AI Studio feedback , prompt	20	3090	January 29, 2026

Gemini 3 significantly worse thant 2.5 Pro at long context. Temperature likely to blame

Related topics