← All models

DeepSeek-Chat on VCBench

DeepSeek-Chat scores F0.5 12.1 on VCBench, the venture capital benchmark from the University of Oxford and Vela Research, with 80.6% precision and 3.0% recall on the 4,500-founder private test set. That is rank 31 of 34, 1.1× the F0.5 of tier-1 VCs and 1.4× Y Combinator, at $0.17 per 1,000 founders scored.

DeepSeek-Chat on the VCBench leaderboard today
Rank
#31 of 34
F0.5
12.1
Precision
80.6%
Recall
3.0%
Cost / 1k founders
$0.17

Scored on the 4,500-founder private test set, mean over three folds. That is 1.1× the F0.5 of tier-1 VCs. See the full leaderboard.

Compared with the reference rows

EntryPrecisionRecallF0.5Cost / 1k
DeepSeek-Chat80.6%3.0%12.1$0.17
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. DeepSeek-Chat used about 284 input and 71 output tokens per founder on deepseek-flash, 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 ($0.3 input and $1.2 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 DeepSeek-Chat score on VCBench?
DeepSeek-Chat scores F0.5 12.1 on VCBench, with 80.6% precision and 3.0% recall, rank 31 of 34 as of 2026-10-26.
Does DeepSeek-Chat beat human investors at predicting founder success?
Yes. Its F0.5 is 1.1 times that of tier-1 VCs (10.7) and 1.4 times Y Combinator (8.6), after both are normalized to the dataset's 9% base rate.
How much does DeepSeek-Chat cost per 1,000 founders?
About $0.17 per 1,000 founders at list prices on 2026-09-24. An LLM reads each founder profile at scoring time and returns a prediction. DeepSeek-Chat used about 284 input and 71 output tokens per founder on deepseek-flash, 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 ($0.3 input and $1.2 output per million tokens), so it recomputes when prices change. Build costs such as question or policy generation are excluded.