Gemini 3.6 Flash

Google's July 21, 2026 "workhorse" refresh of its Flash tier (API gemini-3.6-flash). Intelligence is held flat — AA Intelligence Index v4.1: 50, the same as Gemini 3.5 Flash — while the model gets dramatically more efficient: Google reports it roughly halves time-per-task (1.3 min vs 2.7 min) and uses ~17% fewer output tokens, with output pricing cut to $7.50 / 1M (from $9.00; input unchanged at $1.50). The knowledge cutoff moves up to March 2026.

Multimodal input (text, image, video, audio, PDF), text output, a 1M-token context and 64K max output. Google cites large agentic and coding gains over 3.5 Flash: DeepSWE 49% (vs 37%), MLE-Bench 63.9% (vs 49.7%), OSWorld-Verified 83.0% (vs 78.4%), GDPval-AA v2 1421 (vs 1349). Proprietary and API-only — GA in AI Studio, the Gemini API, Vertex AI / Gemini Enterprise, the Gemini app, Android Studio, and Antigravity.