Will Gemini 3.0 Pro exceed 80% on SWE-Bench verified?
18
1kṀ11kresolved Nov 19
Resolved
NO1H
6H
1D
1W
1M
ALL
Will resolve per Google’s reporting
Update 2025-11-09 (PST) (AI summary of creator comment): The result must be achieved with pass @1 (single attempt) with no parallel test time compute allowed.
This question is managed and resolved by Manifold.
Get 
1,000 to start trading!
1,000🏅 Top traders
| # | Name | Total profit |
|---|---|---|
| 1 | Ṁ471 | |
| 2 | Ṁ152 | |
| 3 | Ṁ141 | |
| 4 | Ṁ123 | |
| 5 | Ṁ87 |
Sort by:
@JaundicedBaboon https://blog.google/products/gemini/gemini-3/#gemini-3-deep-think
They are reporting a score of 76.2%
Does it matter whether or not they report parallel test time compute?
People are also trading
Related questions
Gemini 3.0 Pro outperforms GPT-5 on METR 50% time horizon?
77% chance
Gemini 3 Pro exits preview by EOY?
18% chance
Will Gemini 3.0 Pro Achieve Silver Medal Performance on the 2025 IMO?
49% chance
Will Gemini 3.0 Pro Preview have very low rate limits OR high latency OR multiple unstable weights?
21% chance
Before 2026, will Gemini 3.0 exceed GPT-5 in Metr estimated time horizon?
76% chance
Who will overtake Gemini 2.5 pro when?
How many will Gemini 3.0 achieve? [Read description]
2.19
Will Gemini 3 Pro be better than ChatGPT 5 Thinking in math?
89% chance