Google Lists Gemini 3.8 Flash on OpenRouter With 1M-Token Context, September 2026 Release Date
Google's Gemini 3.8 Flash has surfaced on OpenRouter with a 1-million-token context window and discounted pricing of $0.75 per 1M input tokens and $3.75 per 1M output tokens. Google has not issued a separate public announcement, and the listed release date of September 2, 2026 is unusually far out, leaving key details unconfirmed.
Gemini 3.8 Flash — Quick Specs
Gemini 3.8 Flash Appears on OpenRouter
Google's Gemini 3.8 Flash is now listed on OpenRouter, the model-routing marketplace, with a context window of 1 million tokens and a discounted price of $0.75 per 1M input tokens and $3.75 per 1M output tokens. OpenRouter's listing marks the price as "50% off," implying a standard rate of roughly $1.50 per 1M input tokens and $7.50 per 1M output tokens once the promotion ends.
As of publication, Google has not issued a corresponding blog post, developer changelog, or press release confirming the model independently of the OpenRouter listing. The information in this article is sourced entirely from that listing.
What's Known
According to the OpenRouter product page, Gemini 3.8 Flash is described as "Google's most intelligent Flash model, engineered for long-horizon software engineering, autonomous agents, and complex enterprise workflows." That description is a company/marketplace claim, not an independently verified benchmark result — no MMLU, HumanEval, or other standard benchmark scores were published alongside the listing.
Confirmed specifications from the listing:
- Context window: 1,000,000 tokens
- Input pricing: $0.75 per 1M tokens (listed as 50% off)
- Output pricing: $3.75 per 1M tokens (listed as 50% off)
- Listed release date: September 2, 2026
- OpenRouter uptime (24-hour window): 100%
The September 2, 2026 release date is notable because it falls well in the future relative to typical model-release reporting conventions, and it has not been corroborated by any Google-owned channel. It's possible this is a placeholder date in OpenRouter's system, a scheduling artifact, or a genuine forward-dated listing tied to a staged rollout. Readers should treat the date as unconfirmed until Google publishes its own documentation.
No training cutoff date, parameter count, or modality breakdown (text-only vs. multimodal input/output) was disclosed in the source listing. Given that Gemini models have historically supported text, image, and in some cases audio and video input, Gemini 3.8 Flash is presumed multimodal, but this has not been explicitly confirmed for this specific version.
What This Means
This listing gives builders an early pricing and context-window signal for Gemini 3.8 Flash before any formal Google announcement — a pattern that has become common as model marketplaces like OpenRouter surface new checkpoints ahead of official vendor blog posts. The 1M-token context window keeps pace with Google's Gemini 1.5/2.x Flash line, and the sub-$1 input pricing keeps Flash positioned as the low-cost, high-throughput tier of Gemini's lineup rather than a frontier model competing on raw benchmark scores.
The unresolved release date is the story's biggest open question. Until Google confirms Gemini 3.8 Flash through its own channels — with a verifiable release date, benchmark results, and modality details — developers building on this model via OpenRouter should treat the listing as early access information rather than a finalized product announcement.
Related Articles
Google Launches Gemini 3.8 Flash With Same Pricing as Predecessor, Its Third Flash Model in Six Weeks
Google released Gemini 3.8 Flash, its third Flash model in six weeks, pricing it identically to its predecessor at $0.75 per million input tokens and $3.75 per million output tokens. The launch coincides with a favorable antitrust ruling and public praise from Berkshire Hathaway's Greg Abel, giving Alphabet a stronger narrative after a four-month stock slide.
Meta Releases Muse Spark 1.3, a Free Multimodal Reasoning Model with 1M-Token Context
Meta has released Muse Spark 1.3, a multimodal reasoning model with a 1M-token context window, listed as free on OpenRouter. The model targets long-running agentic, multi-agent, and coding workflows, though audio input support remains incomplete.
Google Launches WeatherNext 3, Claims 50% More Accurate Precipitation Forecasts
Google DeepMind and Google Research released WeatherNext 3, a weather AI model trained on real-time geostationary satellite data instead of lagging numerical weather prediction outputs. Google claims up to 50% more accurate day-ahead precipitation forecasts, now rolling out to Search, Maps, and the Gemini app.
InclusionAI Releases Ling 3.0 Flash Fin, a Finance-Focused MoE Model with 5.1B Active Parameters
InclusionAI has released Ling 3.0 Flash Fin, a finance-specialized mixture-of-experts model built on Ling 3.0 Flash. The model activates 5.1B of its 124B total parameters and targets long-horizon investment planning tasks while retaining general reasoning, coding, and math capabilities.
Comments
Loading...