Tech Meridian← ALL MODELS
PROMY MERIDIAN

GOOGLE DEEPMIND · MODEL RELEASE TRACKER

Gemini 3.8 Flash

Gemini 3.8 Flash is a named variant of Google's Gemini 3.8 series announced in September 2026. It was listed among Google’s September AI releases and has been used in external research (via OpenRouter).

CURRENT SNAPSHOT4/5 DIMENSIONS WITH DATA

The dimensions that change the decision.

PRICING
Reported cached-input discountReported to have a 90% cache discount on cached input tokens (90 percent cheaper cached inputs).
CONTEXT WINDOW

Not established from the available sources.

MODALITIES
Modality — multimodal (visual, audio)Described as multimodal; used in visual projects (e.g., image transformations and multi-camera prototypes) and noted alongside new audio models released the same month.
BENCHMARKS
Benchmark / relative performanceReports substantial gains over Gemini 3.7 Flash and often approaches the performance of higher-cost frontier models.
AVAILABILITY
Observed availability in researchUsed in a research experiment via OpenRouter (the arXiv paper reports using “Gemini 3.8 Flash via OpenRouter” in the Pi agent harness).

RELEASE TIMELINE

Published, source-backed release events only.

5 SOURCES · IMPORTANCE 9.0

Google DeepMind unveils Gemini 4 Argon, a frontier model with 1M-token context

Google DeepMind announced Gemini 4 Argon, a new frontier multimodal model optimized for long-horizon reasoning and complex workflows. Argon is rolling out initially to trusted cyber defenders via the Fairwind Program, expands context length to 1 million tokens, reports leading benchmark performance across coding, finance, legal, and video understanding, and will be made more widely available after phased safety testing and engagement with U.S. government pre-release processes; Google also published introductory pricing.

→

VERIFIABLE FACTS

Every value stays attached to a source and date.

CAPABILITIES · Capabilities — advanced reasoning and agentic tasksDEVELOPER CLAIM

Delivers substantial gains in advanced, multistep reasoning and agentic tasks (including software engineering), improving over 3.7 Flash by executing extra reasoning steps and iteratively calling tools to produce more accurate outputs.

LIMITATIONS · Limitation observed with long profession-specific promptsINDEPENDENTLY SUPPORTED

In a published evaluation, loading full profession-specific profiles with Gemini 3.8 Flash produced 1.5–2.3× as many output tokens and increased estimated cost per successful call by 2.2–4.5×. On 60 tool-using BioMysteryBench problems, mean solve rates were 46.7% with the profile vs. 56.7% at baseline (difference −10.0 percentage points), driven by more frequent token- and time-limit stops under the profile for the tested model.

WHAT CHANGED

Stored passport versions, without reconstructed history.

Passport updated8 facts
RELEASE · Announcement / releaseIntroduced/launched in September 2026.→Announced by Google in September 2026 as part of September AI updates (listed alongside other Gemini 3.8 variants).
AVAILABILITY · Availability− Available across Google products and tools (examples given: Google Antigravity and Google AI Studio); builders have been using it since launch.
PRICING · Reported cached-input discount+ Reported to have a 90% cache discount on cached input tokens (90 percent cheaper cached inputs).
LIMITATIONS · Limitation observed with long profession-specific prompts+ In a published evaluation, loading full profession-specific profiles with Gemini 3.8 Flash produced 1.5–2.3× as many output tokens and increased estimated cost per successful call by 2.2–4.5×. On 60 tool-using BioMysteryBench problems, mean solve rates were 46.7% with the profile vs. 56.7% at baseline (difference −10.0 percentage points), driven by more frequent token- and time-limit stops under the profile for the tested model.
AVAILABILITY · Observed availability in research+ Used in a research experiment via OpenRouter (the arXiv paper reports using “Gemini 3.8 Flash via OpenRouter” in the Pi agent harness).
Passport created6 facts