AIEvals

Sources

huggingface.co

This reference reads 2 published tables · 1 benchmark page from huggingface.co. Each source below links to the original publication and the pages it supports.

Published lab tablesModel pagesBenchmark sites
Pages read
3
Published figures
175
Last read
17 October 2026

Pages read from this site

Original publications, their figure counts, and where they are used in this reference.

  1. DeepSeek: DeepSeek-V4.1-Flash-1017: complete release comparison

    huggingface.co/deepseek-ai/DeepSeek-V4.1-Flash-1017
    65figures
    Benchmark siteLab table published 17 October 2026DeepSeek dated model card and complete evaluation table

    Used on these pages

    Show 10 more linked pages

  2. DeepSeek: DeepSeek-V4.1-Flash: comparison with closed and open-source models

    huggingface.co/deepseek-ai/DeepSeek-V4.1-Flash/resolve/main/DeepSeek_V41_Tech_Report.pdf
    110figures
    Benchmark siteLab table published 10 September 2026

    Used on these pages

    Show 15 more linked pages

  3. MMLU Pro

    huggingface.co/datasets/TIGER-Lab/MMLU-Pro
    —figures
    Benchmark site

    Used on these pages

Counts show published figures attributed to each source. A dash means no model scores are counted here.