Skip to content
aiqbench

DeepSeek · model passport

DeepSeek: DeepSeek V4 Pro 0813

deepseek-v4-pro-0813
  • Open weights
  • 1.65T params
  • 1,048,576 context
  • Released 2026-08-13(HF repo created)
4/27axes covered
Best percentile
89.5AIME (math)
Cohort standing
67.3median 59.8

What the data says

Benchmarked on 4 of 27 tracked axes. Every figure below is the model's own published result on that board.

Strongest axes

  • 89.5 pct — AIME (math) (98.6%)
  • 88.7 pct — GPQA Diamond (91.7%)

Uneven: 19.8 percentile points between its best and worst axis — 89.5 pct on AIME (math) versus 69.7 pct on ARC-AGI-2.

  • AIME (math)89.5 pct98.6%Verify →
  • GPQA Diamond88.7 pct91.7%Verify →
  • SimpleQA Verified73.2 pct52.9%Verify →
  • ARC-AGI-269.7 pct61.3%Verify →
  • OverallNot benchmarked
  • WebDev ArenaNot benchmarked
  • Vision ArenaNot benchmarked
  • Factuality ArenaNot benchmarked
  • Search ArenaNot benchmarked
  • Document ArenaNot benchmarked
  • Agent ArenaNot benchmarked
  • FrontierMathNot benchmarked
  • Humanity's Last ExamNot benchmarked
  • SimpleBenchNot benchmarked
  • LiveBenchNot benchmarked
  • BFCL v4 (function calling)Not benchmarked
  • SWE-bench Verified (live board)Not benchmarked
  • Terminal-Bench 4.0Not benchmarked
  • MMLU-Pro (Artificial Analysis)Not benchmarked
  • LiveCodeBench (Artificial Analysis)Not benchmarked
  • IFBench (Artificial Analysis)Not benchmarked
  • HHEM Hallucination (Vectara)Not benchmarked
  • SWE-bench VerifiedNot benchmarked
  • Terminal-BenchNot benchmarked
  • CodingNot benchmarked
  • Instruction followingNot benchmarked
  • Hard promptsNot 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-V4-Pro-0813
Parameters1.65T
ArchitectureDeepseekV4ForCausalLM

Transformers

from transformers import AutoModelForCausalLM, AutoTokenizer

model_id = "deepseek-ai/DeepSeek-V4-Pro-0813"
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 →

Use it hosted

API model IDdeepseek/deepseek-v4-pro-0813
Context length1,048,576 tokens
Max output384,000 tokens
Input price$0.66 per 1M tokens
Cached input price$0.02 per 1M tokens
Output price$1.98 per 1M tokens

API parameters this endpoint supports

Frequency penaltyInclude reasoningLogit biasLogprobsMax tokensMin-pPresence penaltyReasoningReasoning effortRepetition penaltyResponse formatSeedStop sequencesStructured outputsTemperatureTool choiceTool callingTop-kTop logprobsTop-p

Example request

curl https://openrouter.ai/api/v1/chat/completions \
  -H "Authorization: Bearer $OPENROUTER_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"deepseek/deepseek-v4-pro-0813","messages":[{"role":"user","content":"Hello"}]}'

View on OpenRouter →

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-v4-pro-0813 stands at 67.3 (raw mean 80.3) — measured on 4 of 27 axes.

    standingtemplateVerify →

  • deepseek-v4-pro-0813 stands at 67.3 versus a cohort median of 59.8 across 188 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 →