Google DeepMind Releases Gemini 3.8 Flash AI Model with Enhanced Reasoning Capabilities

Google's DeepMind has released Gemini 3.8 Flash, its fourth Flash model in under four months. This new model scores 59 on the Artificial Analysis Intelligence Index, a 3-point improvement over its predecessor and comparable to sub-maximum reasoning efforts of GPT-5.6 Sol (xhigh, 59) and Grok 4.6 (medium, 59).

The Gemini 3.8 Flash is priced at $0.75/$3.75 per million input/output tokens until the end of the year, matching the discounted pricing of its predecessor. However, despite unchanged per-token pricing, the new model costs ~40% more than Gemini 3.7 Flash due to a 30% increase in average output tokens per task to 48k.

The model's high reasoning level achieves the Intelligence vs. Cost per Task Pareto frontier at $0.58 per task, comparable to GPT-5.6 Terra (max, $0.53). With medium reasoning, Gemini 3.8 Flash costs $0.41 per task, while with low reasoning it falls to $0.24.