DeepSeek-V2

By DeepSeek, in the DeepSeek-V family.

Category scores

Out of 100, from independent and third-party results only, by method category-2026-09-29.1.

Not enough independent results yet

Only independent and third-party results rank a model; self-reported and unattributed numbers never do.

Every result we hold for this model, with who ran it and where it was published.

Knowledge & reasoning results
Benchmark	Metric	Value	Provenance	Setting	Evaluated	Source
AI2 Reasoning Challenge (ARC)	Challenge set accuracy	92.2%	Self-reported	—	—	DeepSeek-V3 Technical Report via Epoch AI
BIG-Bench Hard (BBH)	Average accuracy	78.8%	Self-reported	—	—	DeepSeek-V3 Technical Report via Epoch AI
BIG-Bench Hard (BBH)	Average accuracy	78.8%	Self-reported	—	—	DeepSeek-V3 Technical Report via Epoch AI
HellaSwag	Accuracy	87.1%	Self-reported	—	—	DeepSeek-V3 Technical Report via Epoch AI
MMLU	Accuracy	78.4%	Self-reported	—	—	DeepSeek-V3 Technical Report via Epoch AI
PIQA	Accuracy	83.9%	Self-reported	—	—	DeepSeek-V3 Technical Report via Epoch AI
WinoGrande	Accuracy	86.3%	Self-reported	—	—	DeepSeek-V3 Technical Report via Epoch AI

Factuality & hallucination results
Benchmark	Metric	Value	Provenance	Setting	Evaluated	Source
TriviaQA	Exact match	80.0%	Self-reported	—	—	DeepSeek-V3 Technical Report via Epoch AI