The most prestigious law school admissions discussion board in the world.
|
 |
 
|
Gemini 3 Pro and NotebookLM will revolutionize the practice of shitlaw
| scholarship | 11/19/25 |
| scholarship | 11/19/25 |
| ,.,...,.,.,...,.,,. | 11/19/25 |
| scholarship | 11/19/25 |
| scholarship | 11/19/25 |
Poast new message in this thread
Date: November 19th, 2025 8:58 PM Author: scholarship (http://www.autoadmit.com/thread.php?thread_id=5800126&forum_id=2],#49445142) |
Date: November 19th, 2025 9:50 PM Author: scholarship Gemini 3 Pro scored 87.9% on BigLaw Bench, an improvement over Gemini 2.5 Pro (85%) and in line with other frontier models. It particularly excelled at transactional work and in the litigation drafting sub-tasks, with better tone control, stylistic polish, and creative enhancements to content when appropriate.
https://www.harvey.ai/blog/gemini-3-pro-public-preview-early-access-evaluation-results |
 |
Date: November 19th, 2025 10:50 PM Author: ,.,...,.,.,...,.,,. That’s a surprisingly small jump given the performance increase on other benchmarks from 2.5 pro (http://www.autoadmit.com/thread.php?thread_id=5800126&forum_id=2],#49445305) |
|
Date: November 19th, 2025 10:52 PM Author: scholarship should’ve been named 2.6 (http://www.autoadmit.com/thread.php?thread_id=5800126&forum_id=2],#49445310) |
|
Date: November 19th, 2025 9:57 PM Author: scholarship GPT-5.1 still smokes it on the benchmark if they’re using the same thing:
On our BigLaw Bench evaluation suite, GPT-5.1 achieved the highest score yet for the GPT family of models at 91.8%. This underscores its potential to handle the rigor and nuance of elite legal practice.
https://www.harvey.ai/blog/gpt-5-1-now-live-in-harvey |