← All models

Claude 3.5 Haiku on VCBench

Claude 3.5 Haiku scores F0.5 18.2 on VCBench, the venture capital benchmark from the University of Oxford and Vela Research, with 15.8% precision and 46.4% recall on the 4,500-founder private test set. That is rank 28 of 34, 1.7× the F0.5 of tier-1 VCs and 2.1× Y Combinator, at $0.52 per 1,000 founders scored.

Claude 3.5 Haiku on the VCBench leaderboard today
Rank
#28 of 34
F0.5
18.2
Precision
15.8%
Recall
46.4%
Cost / 1k founders
$0.52

Scored on the 4,500-founder private test set, mean over three folds. That is 1.7× the F0.5 of tier-1 VCs. See the full leaderboard.

Compared with the reference rows

EntryPrecisionRecallF0.5Cost / 1k
Claude 3.5 Haiku15.8%46.4%18.2$0.52
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. Claude 3.5 Haiku used about 284 input and 74 output tokens per founder on claude-3-5-haiku, 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 ($0.8 input and $4 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 Claude 3.5 Haiku score on VCBench?
Claude 3.5 Haiku scores F0.5 18.2 on VCBench, with 15.8% precision and 46.4% recall, rank 28 of 34 as of 2026-10-26.
Does Claude 3.5 Haiku beat human investors at predicting founder success?
Yes. Its F0.5 is 1.7 times that of tier-1 VCs (10.7) and 2.1 times Y Combinator (8.6), after both are normalized to the dataset's 9% base rate.
How much does Claude 3.5 Haiku cost per 1,000 founders?
About $0.52 per 1,000 founders at list prices on 2026-09-24. An LLM reads each founder profile at scoring time and returns a prediction. Claude 3.5 Haiku used about 284 input and 74 output tokens per founder on claude-3-5-haiku, 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 ($0.8 input and $4 output per million tokens), so it recomputes when prices change. Build costs such as question or policy generation are excluded.