Skip to content
aiqbench

DeepSeek · model passport

DeepSeek V3 0324

deepseek-v3-0324
  • Open weights
  • MIT licence
  • 684.5B params
  • Released 2025-03-24(curated)
11/27axes covered
Overall · LMArena
1,395#154 of 396
Best percentile
77.9MMLU-Pro (Artificial Analysis)
Cohort standing
52.0median 46.5

What the data says

Benchmarked on 11 of 27 tracked axes. Every figure below is the model's own published result on that board.

Uneven: 57.2 percentile points between its best and worst axis — 77.9 pct on MMLU-Pro (Artificial Analysis) versus 20.7 pct on SimpleBench.

  • MMLU-Pro (Artificial Analysis)77.9 pct81.9%Verify →
  • LiveBench67.7 pct62.8%Verify →
  • Overall61.4 pct1,395Verify →
  • Instruction following59.1 pct1,378Verify →
  • Hard prompts58.8 pct1,408Verify →
  • Coding56.3 pct1,428Verify →
  • GPQA Diamond49.8 pct67.6%Verify →
  • LiveCodeBench (Artificial Analysis)49.6 pct40.5%Verify →
  • IFBench (Artificial Analysis)42.3 pct41.0%Verify →
  • AIME (math)35 pct37.8%Verify →
  • SimpleBench20.7 pct27.2%Verify →
  • WebDev ArenaNot benchmarked
  • Vision ArenaNot benchmarked
  • Factuality ArenaNot benchmarked
  • Search ArenaNot benchmarked
  • Document ArenaNot benchmarked
  • Agent ArenaNot benchmarked
  • FrontierMathNot benchmarked
  • SimpleQA VerifiedNot benchmarked
  • ARC-AGI-2Not benchmarked
  • Humanity's Last ExamNot benchmarked
  • BFCL v4 (function calling)Not benchmarked
  • SWE-bench Verified (live board)Not benchmarked
  • Terminal-Bench 4.0Not benchmarked
  • HHEM Hallucination (Vectara)Not benchmarked
  • SWE-bench VerifiedNot benchmarked
  • Terminal-BenchNot benchmarked

Percentiles are the model's standing within each board's own population. Axes marked "not benchmarked" have no published result for this model — that is an absence of data, not a low score.

How to run it

Run it yourself

Repositorydeepseek-ai/DeepSeek-V3-0324
Parameters684.5B
ArchitectureDeepseekV3ForCausalLM
LicenceMIT

Transformers

from transformers import AutoModelForCausalLM, AutoTokenizer

model_id = "deepseek-ai/DeepSeek-V3-0324"
tokenizer = AutoTokenizer.from_pretrained(model_id)
model = AutoModelForCausalLM.from_pretrained(model_id)

Ollama

We don't sync Ollama tags, so we won't print a command we can't verify — a wrong one would pull the wrong model. Search the Ollama library →

Unsloth fine-tuning recipes are model-specific and we don't derive them here. Unsloth docs →

Identifiers, limits and prices are synced from the upstream source shown above. Any field we haven't synced is left out entirely rather than guessed.

  • deepseek-v3-0324 stands at 52 (raw mean 52.6) — measured on 11 of 27 axes.

    standingtemplateVerify →

  • deepseek-v3-0324 stands at 52 versus a cohort median of 46.5 across 151 models released within 180 days — above the peer median.

    cohorttemplateVerify →

Every claim is a computed fact from the current board data, phrased by a low-temperature model call and rejected by a validation gate if any number is not in the source row. How the leaderboards work →