# Gemini-2.5-pro-preview-06-05 seems worse than gemini-2.5-pro-preview-05-06 in academic writing

**URL:** <https://discuss.ai.google.dev/t/gemini-2-5-pro-preview-06-05-seems-worse-than-gemini-2-5-pro-preview-05-06-in-academic-writing/88460>\
**Category:** Gemini API\
**Tags:** models, gemini-2-5\
**Created:** [June 11, 2025, 11:08pm UTC](https://discuss.ai.google.dev/t/gemini-2-5-pro-preview-06-05-seems-worse-than-gemini-2-5-pro-preview-05-06-in-academic-writing/88460 "2025-06-11T23:08:01Z")\
**Posts on this page:** 3\
**Page:** 1

<div class="post-metadata">

**Author:** ![Xu\_Cui](https://d1dlmcr85iqnpo.cloudfront.net/user_avatar/discuss.ai.google.dev/xu_cui/32/55458_2.png) [@Xu\_Cui](https://discuss.ai.google.dev/u/Xu_Cui)\
**Post date:** [June 11, 2025, 11:08pm UTC](https://discuss.ai.google.dev/t/gemini-2-5-pro-preview-06-05-seems-worse-than-gemini-2-5-pro-preview-05-06-in-academic-writing/88460/1 "2025-06-11T23:08:01Z")

</div>

I received Google’s email today, saying “Gemini 2.5 Pro Preview 05-06 will be discontinued on June 19, 2025”. I wonder if there is a way to still use gemini-2.5-pro-preview-05-06 even after June 19?

We developed an app for helping scientist write academic papers. The output from gemini-2.5-pro-preview-05-06 is very good. However, when we switch to gemini-2.5-pro-preview-06-05, we notice that the length of the output is considerably shorter, and it also fails to include enough references we provided.

---

<div class="post-metadata">

**Author:** ![Kiran\_Sai\_Ramineni](https://d1dlmcr85iqnpo.cloudfront.net/user_avatar/discuss.ai.google.dev/kiran_sai_ramineni/32/8859_2.png) [@Kiran\_Sai\_Ramineni](https://discuss.ai.google.dev/u/Kiran_Sai_Ramineni)\
**Post date:** [June 12, 2025, 5:06am UTC](https://discuss.ai.google.dev/t/gemini-2-5-pro-preview-06-05-seems-worse-than-gemini-2-5-pro-preview-05-06-in-academic-writing/88460/3 "2025-06-12T05:06:02Z")

</div>

Hi @Xu_Cui, Thanks for sharing the feedback. If possible could you please provide more details on the specific task where the 2.5 Pro (05-06) model outperformed the current 2.5 Pro (06-05) model? understanding these performance differences is important for identifying areas to improve in future iterations. Thank You.

---

<div class="post-metadata">

**Author:** ![Sam\_Van\_Herwaarden](https://d1dlmcr85iqnpo.cloudfront.net/user_avatar/discuss.ai.google.dev/sam_van_herwaarden/32/32173_2.png) [@Sam\_Van\_Herwaarden](https://discuss.ai.google.dev/u/Sam_Van_Herwaarden)\
**Post date:** [June 16, 2025, 8:56pm UTC](https://discuss.ai.google.dev/t/gemini-2-5-pro-preview-06-05-seems-worse-than-gemini-2-5-pro-preview-05-06-in-academic-writing/88460/5 "2025-06-16T20:56:31Z")

</div>

I’m also seeing worse performance in my use case - a bot for helping users put together experimental designs. This involves a lot of interacting with an external library and reasoning through outputs. While 05-06 did well on this, adhering well to my instructions, planning its interaction with the library well, and behaving as expected through the API, 06-05 seems to behave a lot less steadily. One thing it started doing is dropping “thoughts” sections into streamed chunks over the OpenAI compatible API (without metadata marking them as thoughts - there is just a word “thoughts” in the beginning of the chunk content), but behavior overall also just seems less predictable/stable.