Google logo

Gemini 3.8 Flash

Google · Released Sep 2, 2026

ProprietarycodingagentsbudgetCompare this model

$0.75

Input price / MTok

#10 of 33

$3.75

Output price / MTok

#10 of 33

1.0M

Context window

#16 of 34

73.7

DeepSWE v1.1

#1 of 1

89.4

Terminal-bench 2.1

#1 of 1

59

OSWorld-2.0

#1 of 1

61.4

Vals Finance Agent v2

#1 of 1

54.9

HLE-Verified

#1 of 1

Summary

Gemini 3.8 Flash is a proprietary language model from Google, released on Sep 2, 2026.

At $0.75 input and $3.75 output per million tokens, it ranks #10 of 33 for input price, placing it mid-range on input cost.

Its context window of 1.0M tokens ranks #16 of 34 for context depth in this dataset.

It targets coding, agents, budget workloads.

Built on the Gemini 3.7 Flash base rather than a new foundation model, per Google's own model card. Tuned specifically for long-horizon coding and autonomous agents. Shipped one day before Claude Fable 5.1 and GPT-6 Astra, in a 48-hour window where three frontier labs released within days of each other; positioned as the cost-efficient option of the three rather than a raw-capability leader. Knowledge cutoff March 2026 (some domains limited to January 2025). Available same-day in GitHub Copilot.

Benchmarks

BenchmarkScoreMax%RankSource
DeepSWE v1.173.710073.7%only entryGoogle DeepMind model card (self-reporte...
Terminal-bench 2.189.410089.4%only entryGoogle DeepMind model card (self-reporte...
OSWorld-2.05910059.0%only entryGoogle DeepMind model card (self-reporte...
Vals Finance Agent v261.410061.4%only entryGoogle DeepMind model card (self-reporte...
HLE-Verified54.910054.9%only entryGoogle DeepMind model card (self-reporte...