DeepSeek · mit

DeepSeek-V3.2-Exp

DeepSeek-V3.2-Exp by DeepSeek appears in 2 sources with Reasoning at 35.69. Best read for Code quality, Coding, High intelligence.

CreatorDeepSeek

Release date2025-09-29

Knowledge cutoffNot published

ContextUnknown

Input priceUnknown

Output priceUnknown

Modalitytext

CountryCN

Metrics

All source-backed metrics

Open in leaderboard

LLM Stats Rank72ranking · source_rank

Reasoning35.69reasoning · index_reasoning

Math33.44math · index_math

Coding21.73coding · index_code

Research1.46research · index_search

Vision15.78multimodal · index_vision

Finance38.71domain · index_finance

Legal38.68domain · index_legal

Healthcare38.44domain · index_healthcare

GPQA79.9 %reasoning · gpqa_score

AIME 202589.3 %math · aime_2025_score

SWE-bench Verified67.8 %coding · swe_bench_verified_score

Code Arena749.69 %coding · coding_arena_score

Humanity Last Exam19.8 %reasoning · hle_score

SimpleQA97.1 %knowledge · simpleqa_score

BrowseComp40.1 %research · browsecomp_score

Terminal Bench37.7 %coding · terminal_bench_score

Parameters685,000,000,000 paramsmodel · params

Arena Rating1,422.53metric

Arena Rank61metric

Vote Count11,937metric

Evidence

Citations and source overlap

Arena AILM Arena text leaderboard (overall) mirrored in the public leaderboard dataset.Retrieved 2026-05-20 LLM StatsPublic full LLM leaderboard advanced explorer snapshot, including ranking, benchmark indexes, pricing, speed, context, license, and model metadata.Retrieved 2026-06-01

FAQ

How should I read this profile?

Treat this as a source-backed model dossier, not an EvalKit-run verification. The public values are replicated from linked sources and kept source-scoped.

Is DeepSeek-V3.2-Exp verified by EvalKit?

No. EvalKit currently shows 0 verified rows until real run evidence exists.

Why can metrics disagree?

Different sources test different tasks, dates, prompts, and aggregation methods. EvalKit keeps those differences visible instead of merging them into a fake universal score.