Spokes.wiki Search About
Software Application ↗ source url updated Tue Jun 30 2026 00:00:00 GMT+0000 (Coordinated Universal Time)

Gemini

google‘s proprietary, closed-weight frontier model family — the premium track opposite the open-weight gemma-4 family. Delivered via API and Google products rather than self-hosted weights (contrast gemma-4‘s Apache-2.0 self-host model; cf. open-weight-models). This page anchors the Gemini line; specific releases are dated snapshots (llm-benchmarks).

Releases tracked here

  • Gemini 3.5 (May 2026) — positioned as “frontier intelligence for agents and coding,” with multi-step workflow capability google-ai-updates-may-2026. Google’s bid at the agent/coding frontier where anthropic (Claude) and OpenAI compete.
  • Gemini Omni (May 2026) — multimodal generation producing video from image + audio + video + text input google-ai-updates-may-2026. Extends the frontier into generative video, distinct from the input-multimodality of the open gemma-4-12b-announcement.
  • Nano Banana 2 Lite (Jun 2026, gemini-3.1-flash-lite-image) — the speed/cost text-to-image tier; ~4 s generations at $0.034/1k images; replaces first-gen Nano Banana gemini-omni-flash-nano-banana-2-lite.
  • Gemini Omni Flash (Jun 2026, gemini-omni-flash-preview) — cost-efficient video generation + conversational editing; $0.10/sec (Veo-3.1-Fast parity), 10 s max; still preview gemini-omni-flash-nano-banana-2-lite. These two are the family’s first generative-media tiers with concrete IDs and per-unit pricing.
  • Gemini for Science (May 2026) — a science-focused tooling line.
  • Q2 2026 lineup (alphabet-q2-2026-earnings) — the family spread across cost tiers and a release pipeline: Gemini 3.6 Flash + 3.5 Flash-Lite (cost-efficient tiers), Gemini 3.5 Flash Cyber (a cybersecurity-specialized variant, “comparable to larger models” — the first domain-specialized Gemini here), Gemini 3.5 Pro in testing, and Gemini 4 pre-training begun. Consumer Omni video-in-app claimed +40% DAU creating videos. All company-reported.

Positioning

Gemini is the closed frontier half of google‘s dual-track strategy: it carries the proprietary llm-benchmarks / llm-api-pricing premium while gemma-4 seeds the open ecosystem and developer mindshare. The May 2026 framing — agents + coding — places Gemini directly against the proprietary reasoning leaders (anthropic, OpenAI) on the axis this wiki’s synthesis watches: whether the frontier premium survives.

google · gemma-4 · llm-benchmarks · llm-api-pricing · open-weight-models · anthropic

Related pages

Links to

Linked from