← All models

Gemini 2.5 Pro on VCBench

Gemini 2.5 Pro scores F0.5 19.9 on VCBench, the venture capital benchmark from the University of Oxford and Vela Research, with 17.1% precision and 58.0% recall on the 4,500-founder private test set. That is rank 26 of 34, 1.9× the F0.5 of tier-1 VCs and 2.3× Y Combinator, at $11.51 per 1,000 founders scored.

Gemini 2.5 Pro on the VCBench leaderboard today
Rank
#26 of 34
F0.5
19.9
Precision
17.1%
Recall
58.0%
Cost / 1k founders
$11.51

Scored on the 4,500-founder private test set, mean over three folds. That is 1.9× the F0.5 of tier-1 VCs. See the full leaderboard.

Compared with the reference rows

EntryPrecisionRecallF0.5Cost / 1k
Gemini 2.5 Pro17.1%58.0%19.9$11.51
Think-Reason-Learn Ensemble40.6%30.1%37.9$0.87
Tier-1 VCs23.0%5.2%10.7n/a
Y Combinator14.0%6.9%8.6n/a
Random Classifier9.0%9.0%9.0$0

Human rows are normalized to the dataset's 9% success rate, so every entry is compared on the same base rate. Costs use list prices on 2026-09-24.

How it was scored

An LLM reads each founder profile at scoring time and returns a prediction. Gemini 2.5 Pro used about 284 input and 1,116 output tokens per founder on gemini-2.5-pro, estimated from the prompt and the saved responses; reasoning models are assumed to think about 1,000 tokens per founder. Cost is tokens times the provider's list price on 2026-09-24 ($1.25 input and $10 output per million tokens), so it recomputes when prices change. Build costs such as question or policy generation are excluded.

Read more

Frequently asked questions

What does Gemini 2.5 Pro score on VCBench?
Gemini 2.5 Pro scores F0.5 19.9 on VCBench, with 17.1% precision and 58.0% recall, rank 26 of 34 as of 2026-10-26.
Does Gemini 2.5 Pro beat human investors at predicting founder success?
Yes. Its F0.5 is 1.9 times that of tier-1 VCs (10.7) and 2.3 times Y Combinator (8.6), after both are normalized to the dataset's 9% base rate.
How much does Gemini 2.5 Pro cost per 1,000 founders?
About $11.51 per 1,000 founders at list prices on 2026-09-24. An LLM reads each founder profile at scoring time and returns a prediction. Gemini 2.5 Pro used about 284 input and 1,116 output tokens per founder on gemini-2.5-pro, estimated from the prompt and the saved responses; reasoning models are assumed to think about 1,000 tokens per founder. Cost is tokens times the provider's list price on 2026-09-24 ($1.25 input and $10 output per million tokens), so it recomputes when prices change. Build costs such as question or policy generation are excluded.