DeepSeek R1 benchmark results
Compare DeepSeek R1 benchmark results across hosted providers and endpoints. This page summarizes public runs on MMLU, MATH, GSM8K, IFEval, and MuSR, including score, latency, sample coverage, prompts, outputs, and methodology.
Provider Endpoints
DeepSeek R1 is tracked on Benchscope. Public evaluation runs will appear here as providers submit results. Provider-hosted versions of the same model can differ in quantization, infrastructure, and serving configuration.
How to Compare Endpoints
Use canonical-prompt runs on the same benchmark to compare endpoints fairly. Score differences between providers running the same model family reflect hosting differences rather than model differences. Check the methodology for how runs are defined and what makes them comparable.
MMLU results for DeepSeek R1
No public MMLU runs for DeepSeek R1 are available yet. Explore all MMLU benchmark results or run this benchmark on your endpoint.
MATH results for DeepSeek R1
No public MATH runs for DeepSeek R1 are available yet. Explore all MATH benchmark results or run this benchmark on your endpoint.
GSM8K results for DeepSeek R1
No public GSM8K runs for DeepSeek R1 are available yet. Explore all GSM8K benchmark results or run this benchmark on your endpoint.
IFEval results for DeepSeek R1
No public IFEval runs for DeepSeek R1 are available yet. Explore all IFEval benchmark results or run this benchmark on your endpoint.
MuSR results for DeepSeek R1
No public MuSR runs for DeepSeek R1 are available yet. Explore all MuSR benchmark results or run this benchmark on your endpoint.
Related
Benchscope is a JavaScript app. If the interactive interface does not load, enable JavaScript or use the links above for the main public sections.