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 V4 Pro 0813?

128 SI-IQ · 90% interval 125–131
Rank 21 of 80 ranked models · Plausible rank range 13–37

DeepSeek V4 Pro 0813 is a model from DeepSeek, released on Aug 13, 2026. Its SI-IQ is 128 (90% interval 125–131), estimated from 8 benchmark results by 2 independent evaluators. Its strongest area is Reasoning and its weakest is Math.

Facts

Developer: DeepSeek
Released: Aug 13, 2026
Access: Open weights (restricted licence)
Benchmarks used: 8 results from 2 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
Chess Puzzles	Reasoning	46.0%	131	Epoch AI	`deepseek-v4-pro-0813_max`	Oct 18, 2026	Epoch AI Benchmarking Hub · data file
Mystery Game Puzzles	Reasoning	43.0%	133	Epoch AI	`deepseek-v4-pro-0813_max`	Oct 18, 2026	Epoch AI Benchmarking Hub · data file
FrontierMath Tier 4	Math	26.8%	124	Epoch AI	`deepseek-v4-pro-0813_max`	Oct 18, 2026	Epoch AI Benchmarking Hub · data file
FrontierMath Tiers 1–3	Math	64.6%	126	Epoch AI	`deepseek-v4-pro-0813_max`	Oct 18, 2026	Epoch AI Benchmarking Hub · data file
OTIS Mock AIME 2024–2025	Math	98.6%	133	Epoch AI	`deepseek-v4-pro-0813_max`	Oct 18, 2026	Epoch AI Benchmarking Hub · data file
GPQA Diamond	Knowledge	91.7%	128	Epoch AI	`deepseek-v4-pro-0813_max`	Oct 18, 2026	Epoch AI Benchmarking Hub · data file
GPQA Diamond (OpenRouter run)	Knowledge	89.3%	128	OpenRouter	`deepseek/deepseek-v4-pro-20260813` BaseTen · median over providers (20 of 20 rows), lower median	Oct 18, 2026	OpenRouter benchmark datasets · data file
SimpleQA Verified	Knowledge	52.9%	132	Epoch AI	`deepseek-v4-pro-0813_max`	Oct 18, 2026	Epoch AI Benchmarking Hub · data file
τ²-bench Airline (OpenRouter run)	Agents	76.8%	132	OpenRouter	`deepseek/deepseek-v4-pro-20260813` Phala · median over providers (26 of 26 rows), lower median	Oct 18, 2026	OpenRouter benchmark datasets · data file

Models with similar SI-IQ

Model	SI-IQ
DeepSeek V4.1 FlashDeepSeek	128	DeepSeek V4 Pro 0813 vs DeepSeek V4.1 Flash
Muse Spark OpenMeta AI	127	DeepSeek V4 Pro 0813 vs Muse Spark Open
Kimi K3Moonshot	127	DeepSeek V4 Pro 0813 vs Kimi K3
Claude Opus 4.7Anthropic	127	DeepSeek V4 Pro 0813 vs Claude Opus 4.7
Gemini 3.1 ProGoogle DeepMind	127	DeepSeek V4 Pro 0813 vs Gemini 3.1 Pro

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

Data sources: Epoch AI, ‘Capabilities & benchmarking’ (CC BY 4.0); Source: OpenRouter (openrouter.ai), licensed under CC BY 4.0: reuse and republish with attribution to OpenRouter..