Wednesday, Jul 22 | --:--
Back to home

Google Ships Gemini 3.6 Flash and 3.5 Flash-Lite; Cyber Tier Stays Limited

On July 21, 2026, Google launched Gemini 3.6 Flash ($1.50/$7.50 per 1M tokens) and 3.5 Flash-Lite ($0.30/$2.50) for agents at scale, while Gemini 3.5 Flash Cyber in CodeMender stays limited to governments and trusted partners; 3.5 Pro remains partner-testing only.

Tech Insights Reporter 6 min read Mountain View, CA
Cover illustration for Google Ships Gemini 3.6 Flash and 3.5 Flash-Lite; Cyber Tier Stays Limited

TLDR

Google on July 21, 2026 introduced Gemini 3.6 Flash, 3.5 Flash-Lite, and a limited 3.5 Flash Cyber pairing with CodeMender. 3.6 Flash is the new workhorse for coding, knowledge work, and multimodal agents at $1.50 / $7.50 per 1M input/output tokens. 3.5 Flash-Lite targets high-throughput agentic tasks at $0.30 / $2.50 and ~350 output tokens/s (Artificial Analysis). 3.5 Pro is still partner testing only; Google also says Gemini 4 pre-training has started.

What shipped

From Google’s Keyword / DeepMind primary (Tulsee Doshi) and DeepMind Flash Cyber post:

Model / product Role Price (per 1M tok) Access day one
3.6 Flash Workhorse: coding, knowledge work, multimodal agents $1.50 in / $7.50 out Gemini API (AI Studio, Android Studio), Antigravity, Gemini Enterprise, Gemini app
3.5 Flash-Lite Fastest 3.5-class; high volume / low latency $0.30 in / $2.50 out Same developer surfaces; also rolling into Google Search
3.5 Flash Cyber + CodeMender Cyber find/fix at scale Lower token price than larger cyber models (not full public list) Governments + trusted partners only (limited pilot)
3.5 Pro Flagship Still not GA — partner testing; “as soon as ready”
Gemini 4 Next gen Most ambitious pre-training run yet started (no ship date)

Vendor-claimed gains for 3.6 Flash vs 3.5 Flash include better DeepSWE (49% vs 37%), MLE Bench (63.9% vs 49.7%), OSWorld-Verified (83.0% vs 78.4%), and GDPval-AA v2 (1421 vs 1349), with computer use as a built-in client-side tool. Google says 3.6 Flash is more token-efficient and cheaper per agentic task than 3.5 Flash despite higher capability. 3.5 Flash-Lite is pitched as beating 3 Flash on several agentic/coding boards (e.g. SWE-Bench Pro 54.2% vs 49.6%; OSWorld-Verified 74.0% vs 65.1%).

Flash Cyber is explicitly dual-use controlled: multi-agent CodeMender stacks for vulnerability discovery and patching, competitive on CyberGym in Google’s charts, not open Chat/API GA.

Independent rankings

Artificial Analysis (launch window):

Model AA Intelligence Index (reported) Notes
Gemini 3.6 Flash (High) ~50 Near GPT‑5.6 Luna Max (~51) and below Terra Xhigh (~52) on contemporaneous comparisons; Google claims ~17% fewer output tokens vs 3.5 Flash on AA token usage for the Index
3.5 Flash-Lite Speed-focused Google cites ~350 output tokens/s on AA; full Intelligence Index still maturing for the new Lite SKU
3.5 Flash (prior) ~55 (High) at earlier board Important baseline: 3.6 Flash prioritizes efficiency/cost and agentic reliability, not necessarily a pure AA score jump over 3.5 Flash High

Public LMArena / Chatbot Arena category Elos for 3.6 Flash were still settling in the first research hours; treat early social Elo screenshots as provisional until vote-stable. Do not invent Arena ranks.

Competitive context

This is Google’s answer to a summer where GPT‑5.6, Grok 4.5, Fable 5, Kimi K3, and Qwen3.8-Max-Preview already own mindshare—while 3.5 Pro remains delayed (see July 16 delay filing). Shipping two Flash SKUs + a defender-only Cyber tier keeps Google in the agent cost/latency race even without flagship Pro GA. Same-day OpenAI cyber-eval breach coverage also makes limited Flash Cyber access a deliberate distribution choice, not a footnote.

Why this story matters

Enterprise agent spend is increasingly Flash-class, not flagship-only. A cheaper, less verbose 3.6 Flash plus a 350 tok/s Lite changes unit economics for ticket routing, document pipelines, and multi-agent coding—while Pro delay + Cyber lock show Google still managing capability and dual-use risk carefully. Buyers should read this as a production stack refresh, not the delayed Pro story ending.

Sources

Featured Image Alt Text

Engraved Gemini hexagonal mark with Flash speed lines and agent workflow lattice for the July 21 3.6 Flash and Flash-Lite launch.

Tags

Google, DeepMind, Gemini 3.6 Flash, Flash-Lite, Flash Cyber, Models, Agents, Artificial Analysis, July 21

Prior Coverage

Earlier Times of AI reporting on this thread.

Scroll to continue reading