Gemini 3.8 Flash
Google · 3 thinking configurations
Intelligence40.9
Input / 1M$0.75
Output / 1M$3.75
Context1M
Tariff: Google · 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 cost2 measured configurations · 1 without both values
Selected: thinking high · Click a point to switch configuration
Gemini 3.8 Flash is Google's most intelligent Flash model with significant gains from 3.7 Flash across software engineering, agentic tasks, and multi-step reasoning.
Intelligence IndexArtificial Analysis composite index, not a percentage. Different versions of the index are not directly comparable.
40.9 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.
73.8 % ±1.4GPQA DiamondGraduate-level scientific questions. Measures specialist reasoning rather than web-search quality.
95.3 %Humanity’s Last ExamDifficult questions across many disciplines, compared under the same evaluation setup.
47.8 %SciCodeCode generation for scientific research problems.
56.6 %Terminal-Bench 2.1Agent tasks in a terminal environment. Results also depend on the evaluation harness.
87.6 %MMMU-ProMultimodal understanding of academic problems that include images.
85.6 %Omniscience accuracyFactual accuracy in Artificial Analysis's Omniscience evaluation.
54.6 %Specifications from the catalog
Inputtext, image, video, file, audio
Outputtext
Tool callingYes
Structured outputYes
Speed238.7 tok/s
First token19.31 s
Maximum output65.5K
Catalog listing date2026-09-02
DeepSWE v1.1 · 113 tasks · mini-swe-agent · cost $2.36 / task · 166 mean steps · 143.2K mean output tokens. Published by the source 9/22/2026.