GoogleGoogle·text

Gemini 3.7 Flash

AudioVisionReasoningWeb searchFunction calling
Advertised as gemini-3-7-flashgemini-3.7-flash
Quick reference
Gemini 3.7 Flash — TLDR
  • - 🆕 Google's newest Flash-tier Gemini, released August 2026
  • - 📏 One-million-token context window for long documents and codebases
  • - 🧠 Tunable thinking budget to trade latency against answer depth
  • - 👁️ Multimodal input including text, images and audio
  • - 🔧 Function calling, tool use and multi-step agentic execution
  • - 🌐 Web-search grounding available for up-to-date answers
  • - 🏢 Hosted via Gemini API and Gemini Enterprise Agent Platform
  • - 🔒 Safety and capability evaluations published in a DeepMind model card
💰 Best price on AntSeed
$0.120 / $0.60068%
per 1M · cheapest in / out
📏 Context
1M tokens
🐜 Sellers
4
advertising on AntSeed
Provider

Google is an American multinational technology corporation and one of the world's most valuable brands. A subsidiary of parent company Alphabet Inc., Google operates across search, cloud computing, consumer electronics, and artificial intelligence. Its DeepMind and Google…

Explore 17 more models by Google
About this model

Gemini 3.7 Flash is Google's mid-tier "workhorse" entry in the Gemini 3 line, sitting between the deeper-reasoning Pro models and the lighter Flash-Lite tier. Google describes it as its most capable Flash model, built for complex coding, agentic workflows and reliable multi-step execution, with a one-million-token context window and tunable thinking. It accepts multimodal input, including text, images and audio, and supports function calling and web-search grounding.

It follows [[sibling:gemini-3-6-flash|Gemini 3.6 Flash]] by roughly five weeks, continuing an unusually rapid cadence within the same family: [[sibling:gemini-3-flash-preview|Gemini 3 Flash Preview]] arrived in December 2025 and [[sibling:gemini-3-5-flash|Gemini 3.5 Flash]] in May 2026. Across that sequence the headline context length has stayed at one million tokens, with the generational work concentrated on coding, tool use and sustained agentic behaviour rather than on raw context expansion. Google publishes a dedicated model card for 3.7 Flash through DeepMind, covering intended uses and safety evaluations.

Within Google's current lineup, [[sibling:gemini-3-5-flash-lite|Gemini 3.5 Flash-Lite]] remains the lower-latency option for simple high-volume tasks, while [[sibling:gemini-3-1-pro-preview|Gemini 3.1 Pro Preview]] targets harder reasoning problems. Gemini 3.7 Flash is offered as a hosted API model — through the Gemini API and the Gemini Enterprise Agent Platform — rather than as open weights, so parameter counts, quantization and licensing details are not disclosed. Developers can dial the thinking level per request to balance response speed against reasoning depth.

Sources
deepmind.googleGemini 3.7 Flash — Google DeepMind· deepmind.googledocs.cloud.google.comGemini 3.7 Flash | Gemini Enterprise Agent Platform | Google Cloud Documentation· docs.cloud.google.com

This About section is AI-generated from public sources via VeniceStats + Venice inference, with no human editing. It may contain inaccuracies.

Usage on AntSeed
Tokens served
35.53k
input + output
Requests
10
settled calls
Buyers
3
distinct, on this model
Sellers used
2
of 4 advertising
Settled
$0.01
gross USDC, this model
Sellers serving Gemini 3.7 Flash (4)compare on the network explorer →
SellerReputationRoutingInput $/MCached $/MOutput $/MCategoriesAPI

"Best price" and the seller table are live AntSeed catalog data (advertised $/1M tokens — or $ per generated image for unit-billed image models — not settled amounts). Reputation = on-chain trust (0-100). "Routing" = the SDK's default buyer routing (what the VPR desktop app ships with): a trust ≥ 60 gate on the effective reputation, then cheapest-first among routable sellers; live failover state (per-peer cooldowns) is buyer-side runtime and not included. Model knowledge (TLDR, provider, About) via the VeniceStats enrichment layer. Advertised catalog, not the model used in any specific purchase. "Usage on AntSeed" counts only settlements whose buyers share the per-model split on-chain (metadata v2/v3, opt-in), so every usage figure is a lower bound.