DeepSeek V4 Flash Vision
DeepSeek · 1 configuration
Intelligence34.8
Input / 1M$0.44
Output / 1M$1.32
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 · 0 without both values
Selected: thinking max · Click a point to switch configuration
Intelligence IndexArtificial Analysis composite index, not a percentage. Different versions of the index are not directly comparable.
34.8 pointsGPQA DiamondGraduate-level scientific questions. Measures specialist reasoning rather than web-search quality.
91.3 %Humanity’s Last ExamDifficult questions across many disciplines, compared under the same evaluation setup.
34.5 %SciCodeCode generation for scientific research problems.
49.7 %Terminal-Bench 2.1Agent tasks in a terminal environment. Results also depend on the evaluation harness.
74.2 %MMMU-ProMultimodal understanding of academic problems that include images.
74.8 %Omniscience accuracyFactual accuracy in Artificial Analysis's Omniscience evaluation.
38.6 %Specifications from the catalog
Inputtext
Outputtext
Tool callingNot available
Structured outputNot available
Speed222.4 tok/s
First token1.01 s
Maximum output—
Catalog listing dateNot available