DeepSeek V4 Flash
DeepSeek · 3 thinking configurations
Intelligence24.2
Input / 1M$0.13
Output / 1M$0.28
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 max · Click a point to switch configuration
Fast DeepSeek V4 lane for economical reasoning, coding, and long-context work
Intelligence IndexArtificial Analysis composite index, not a percentage. Different versions of the index are not directly comparable.
24.2 pointsDeepSWE · pass@1Software-engineering tasks on mini-swe-agent v1.1. Pass@1 across evaluated runs, with measured cost per task. The 95% intervals may overlap.
53.3 % ±3.6GPQA DiamondGraduate-level scientific questions. Measures specialist reasoning rather than web-search quality.
89.4 %Humanity’s Last ExamDifficult questions across many disciplines, compared under the same evaluation setup.
34.8 %SciCodeCode generation for scientific research problems.
45.3 %Terminal-Bench 2.1Agent tasks in a terminal environment. Results also depend on the evaluation harness.
61.8 %τ²-BenchTool use and interaction in assistance scenarios. Does not replace an evaluation of your integration.
95.0 %IFBenchVerifiable instruction following. A proxy for precision rather than a benchmark of creative style.
79.2 %Omniscience accuracyFactual accuracy in Artificial Analysis's Omniscience evaluation.
36.8 %Specifications from the catalog
Inputtext
Outputtext
Tool callingYes
Structured outputYes
SpeedNot available
First tokenNot available
Maximum output16.4K
Catalog listing date2026-04-24
DeepSWE v1.1 · 113 tasks · mini-swe-agent · cost $0.10 / task · 153 mean steps · 107.7K mean output tokens. Published by the source 9/22/2026.