Automated page. Written and translated automatically by AI and checked against the sources by software; no human editor. SI-IQ is an index built from public benchmark results, not an IQ test. How it works

How smart is DeepSeek-V3.2-Exp?

112 SI-IQ · 90% interval 103–121
Not ranked: too little data

DeepSeek-V3.2-Exp is a model from DeepSeek, released on Sep 29, 2025. Its SI-IQ is 112 (90% interval 103–121), estimated from 4 benchmark results by 3 evaluators.

Data as of Oct 18, 2026 · method 1.0.0

A model is ranked when its results come from at least 5 benchmarks (the two GPQA Diamond runs count as one) in at least 3 areas, from at least 2 different evaluators, and its versions are released no more than 60 days apart. This model does not meet that yet, so it is shown with a provisional estimate but not ranked.

Facts

Developer: DeepSeek
Released: Sep 29, 2025
Access: Open weights (restricted licence)
Benchmarks used: 4 benchmark results from 3 evaluators

Ability by area (SI-IQ, 90% interval)

Every benchmark result we used

Each score is one value exactly as the source published it: for Epoch AI the best of its run settings, for OpenRouter the median hosting provider (the developer's own hosting when it ran the test). “Implied SI-IQ” is the ability that this single result points to; the overall SI-IQ is their precision-weighted average.

Benchmark	Area	Score	Implied SI-IQ	Evaluated by	As listed by the source	Retrieved	Source
Aider Polyglot	Coding	74.2%	111	Aider	`DeepSeek-V3.2-Exp (Reasoner)`	Oct 18, 2026	Aider LLM Leaderboards · data file
GPQA Diamond (OpenRouter run)	Knowledge	82.7%	115	OpenRouter	`deepseek/deepseek-v3.2-exp` Novita · median over providers (3 of 3 rows)	Oct 18, 2026	OpenRouter benchmark datasets · data file
Berkeley Function Calling Leaderboard	Agents	56.7%	106	UC Berkeley (Gorilla)	`DeepSeek-V3.2-Exp (Prompt + Thinking)`	Oct 18, 2026	Berkeley Function Calling Leaderboard · data file
τ²-bench Airline (OpenRouter run)	Agents	69.3%	119	OpenRouter	`deepseek/deepseek-v3.2-exp` SiliconFlow · median over providers (2 of 2 rows), lower median	Oct 18, 2026	OpenRouter benchmark datasets · data file

Human preference (Arena)

On the Arena leaderboard, where people vote between two anonymous answers, DeepSeek-V3.2-Exp has a rating of 1,422 (1,415–1,428) from 12,004 votes (published Oct 15, 2026). Human preference is a different measure from benchmark ability and is not part of SI-IQ. Arena (Hugging Face)

Computed with SI-IQ method version 1.0.0. Read the full methodology

Data sources: Aider polyglot leaderboard data (Apache-2.0); Arena leaderboard dataset on Hugging Face (CC BY 4.0); Berkeley Function Calling Leaderboard, Gorilla project (Apache-2.0); Source: OpenRouter (openrouter.ai), licensed under CC BY 4.0: reuse and republish with attribution to OpenRouter (licences and changes).