Google

Google brings the deepest infrastructure advantage to the AI race. Their Gemini 2.5 Pro offers a 1 million token context window, and their Flash models deliver some of the lowest per-token pricing available. At Google I/O 2026 on May 19, Google shipped Gemini 3.5 Flash, the first Flash-tier release that beats the previous Pro flagship on agentic coding suites at roughly 4x the throughput, alongside Gemini Spark, a general-purpose agent that reasons across connected apps. At Google Cloud Next '26 in Las Vegas on April 22, 2026, Google launched the Gemini Enterprise Agent Platform (the evolution of Vertex AI), backed by a $750 million partner fund, and announced that Gemini will power the next generation of Apple's Siri. On July 21, 2026 Google shipped three models in a single announcement: Gemini 3.6 Flash at $1.50/$7.50 per 1M tokens (a cut from the $9.00 output rate on 3.5 Flash), Gemini 3.5 Flash-Lite at $0.30/$2.50 and 350 output tokens per second, and Gemini 3.5 Flash Cyber, a vulnerability-finding model restricted to governments and trusted partners through the CodeMender agent. The framing was efficiency rather than capability: Artificial Analysis scored 3.6 Flash flat at 50 on its Intelligence Index, identical to 3.5 Flash, while measured time per task fell from 2.7 minutes to 1.3 and cost per task from $0.59 to $0.50. The same post confirmed that Gemini 3.5 Pro is still only testing with partners after missing its July 17 target, and that pre-training has begun on Gemini 4, which Google calls its most ambitious run yet. Backed by custom TPU hardware and decades of ML research (Transformer architecture was invented at Google), they compete on both the frontier and the budget ends of the market.