Google's Gemini 3.5 Pro flagship remains unshipped three months past its promised June date, driving Alphabet's worst stock losing streak since 2015.
Google's Gemini 3.5 Pro flagship remains unshipped three months past its promised June date, driving Alphabet's worst stock losing streak since 2015.

Google shipped three Flash models in six weeks while its flagship Gemini 3.5 Pro remains unshipped three months past the June date CEO Sundar Pichai promised, deepening doubts about Alphabet's ability to compete at the AI frontier.
"Gemini 3.8 Flash delivers substantial gains from 3.7 Flash, often approaching the performance of higher-cost frontier models," said Tulsee Doshi, Google senior director of product management, in a blog post announcing the release.
The latest Flash iteration scores 59 on the Artificial Analysis Intelligence Index, up three points from its predecessor and level with GPT-5.6 Sol and Grok 4.6 at comparable reasoning settings. At $0.58 per Intelligence Index task, it is the cheapest model at its level of intelligence, though Anthropic's Claude Fable 5.1 leads the rankings at 66.
The cadence gap carries a price tag. Alphabet shares are on their worst losing streak since 2015 as investors weigh whether Google can hold its position against OpenAI, Anthropic, and Chinese labs shipping open-weight models that have overshadowed Gemini 3.5 Flash on intelligence benchmarks.
The Pro line has not moved since early 2026, making the cheaper Flash tier two versions ahead of the flagship, TNW reported. Google reportedly delayed Gemini 3.5 Pro when its coding performance could not match rival models, according to Ars Technica. The setback follows the departure of Google DeepMind CEO Demis Hassabis in early August, who stepped down to become chairman of the AI research unit, with Koray Kavukcuoglu taking over as SVP.
Pichai's reassurances in a company blog post "weren't particularly convincing," The Register wrote, in light of the missed June deadline. The modest benchmark metrics of Gemini 3.5 Flash have also been overshadowed by open-weight models from Chinese AI companies over the past few months.
Flash Cadence vs. Frontier Gap
Google's rapid Flash iteration is a stopgap while the Pro line catches up. Gemini 3.8 Flash tops the DeepSWE leaderboard for autonomous software engineering, outperforming most larger frontier models at a fraction of the cost, according to Google. The model also shows gains on enterprise benchmarks including Vals Finance Agent V2 and Harvey's Legal Agent Benchmark.
But the Flash line has limits. On OSWorld-2.0, a test of agentic computer use, Gemini 3.8 Flash remains far behind market leader Claude Opus, according to Ars Technica. Google's own figures put Flash behind Claude Opus on agentic computer use, a year after the computer use tool was added at Gemini 3.5.
Pricing for Gemini 3.8 Flash is introductory at $0.75 per million input tokens and $3.75 per million output tokens, doubling to $1.50 and $7.50 in the new year. Rivals have been cutting token prices to keep wary businesses engaged, The Register noted.
What the Delay Costs Alphabet
The stock market has already priced in the execution risk. Alphabet's losing streak — its worst since 2015 — reflects investor concern that Google's AI leadership is slipping. The company faces competition not just from OpenAI and Anthropic at the frontier, but from Chinese labs like GLM and Kimi whose open-weight models now score 60 on the Artificial Analysis Intelligence Index, one point above Gemini 3.8 Flash.
Google has also launched Gemini 3.8 Flash Cyber, a variant tuned for vulnerability detection, but it is limited to trusted testers and governments. The Chrome security team reported a 2.6x improvement in patch accuracy with the new model, and Google's cloud team says Flash Cyber found a critical vulnerability in two hours.
For investors, the question is whether Google can close the gap on its Pro line before the Flash strategy loses credibility. The company's next opportunity to reset expectations comes with the rumored release of a new Gemini coding model this week, CNBC reported. If that too is a Flash variant rather than the long-promised Pro flagship, Alphabet's cadence problem will only deepen.
This article is for informational purposes only and does not constitute investment advice.