DeepSeek V4 Pro
DeepSeek · 3 thinking configurations
Intelligence30.1
Input / 1M$0.44
Output / 1M$0.87
Context1M
Tariff: DeepSeek · Source: artificial-analysis · USD per million tokens
Thinking configuration
What does more thinking buy you?
Compare this model’s measured thinking configurations.
Scroll the chart ↔ · tap a point for details
Efficient frontier: best measured score at each cost1 measured configurations · 2 without both values
Selected: thinking high · Click a point to switch configuration
Open MoE flagship with million-token context for coding and long agent runs
Intelligence IndexArtificial Analysis composite index, not a percentage. Different versions of the index are not directly comparable.
30.1 pointsGPQA DiamondGraduate-level scientific questions. Measures specialist reasoning rather than web-search quality.
90.5 %Humanity’s Last ExamDifficult questions across many disciplines, compared under the same evaluation setup.
35.2 %Terminal-Bench 2.1Agent tasks in a terminal environment. Results also depend on the evaluation harness.
64.8 %τ²-BenchTool use and interaction in assistance scenarios. Does not replace an evaluation of your integration.
94.2 %IFBenchVerifiable instruction following. A proxy for precision rather than a benchmark of creative style.
71.3 %Omniscience accuracyFactual accuracy in Artificial Analysis's Omniscience evaluation.
41.4 %Specifications from the catalog
Inputtext
Outputtext
Tool callingYes
Structured outputYes
Speed85.5 tok/s
First token1.68 s
Maximum output16.4K
Catalog listing date2026-04-24