Anvat
Google1M (128K with full long-context recall) contextgemini-3-5-flash

Gemini 3.5 Flash

Google's fast tier — beats Gemini 3.1 Pro on agentic + coding benchmarks at $0.30/$2.50

Gemini 3.5 Flash shipped at Google I/O 2026 (May 19, 2026) and immediately outperformed the previous-generation Gemini 3.1 Pro on most coding and agentic benchmarks. Terminal-Bench 2.1 at 76.2%, MCP Atlas at 83.6%, GDPval-AA Elo 1656. At $0.30/$2.50 per MTok and 4× faster than other frontier models, it's the strongest cost-quality position in the lineup for agentic work.

Pricing

RateList priceAnvat effectiveSavings
Input$0.30$0.2130%
Output$2.50$1.7530%
Cache read$0.08$0.0529%
All prices per million tokens (MTok). List = provider direct. Anvat effective = 30% discount applied.

Pricing verified 2026-06 · See full Anvat pricing

Strengths

  • Beats Gemini 3.1 Pro on Terminal-Bench (76.2% vs 70.3%) and MCP Atlas (83.6% vs 78.2%)
  • Strongest GDPval-AA Elo of any Gemini Flash (1656)
  • 4× faster output token throughput than other frontier models
  • Cheapest frontier-tier model from the major labs at $0.30/MTok input
  • Native multimodal (text, image, audio, video) in a single model
  • Default model in Google Antigravity 2.0

Where it underperforms

  • Regressed vs Gemini 3.1 Pro on Humanity's Last Exam (40.2% vs 44.4%)
  • Regressed on ARC-AGI-2 (72.1% vs 77.1%)
  • Regressed on MRCR v2 128K long-context recall (77.3% vs 84.9%)
  • Hardest reasoning workloads should wait for Gemini 3.5 Pro (June 2026)

Use cases this model is the right pick for

  • High-throughput agentic loops where speed + cost matter
  • Tool-use / MCP-heavy workflows (top of the field on MCP Atlas)
  • Cursor / Antigravity inline completion at frontier quality
  • Multimodal classification + extraction at volume
  • First-pass router target with escalation to Gemini 3.5 Pro / Opus 4.7

Benchmarks

  • coding

    Terminal-Bench 2.1

    76.2% (vs 70.3% for Gemini 3.1 Pro)

  • agentic

    MCP Atlas

    83.6% (vs 78.2% for Gemini 3.1 Pro)

  • agentic

    GDPval-AA (Elo)

    1656 (vs 1314 for Gemini 3.1 Pro)

  • agentic

    Finance Agent v2

    57.9% (vs 43.0% for Gemini 3.1 Pro)

  • vision

    CharXiv Reasoning

    84.2% (multimodal lead)

  • agentic

    OSWorld-Verified

    78.4%

Benchmark numbers self-reported by provider; verify against the latest publisher documentation before quoting.

Quickstart

Same wire format as direct provider APIs — your existing SDK code keeps working. Point at api.anvat.app/v1 and use your Anvat key.

import OpenAI from "openai";

const client = new OpenAI({
  baseURL: "https://api.anvat.app/v1",
  apiKey: process.env.ANVAT_API_KEY,
});

const response = await client.chat.completions.create({
  model: "gemini-3-5-flash",
  max_tokens: 2048,
  messages: [
    { role: "user", content: "Walk through this codebase and propose a refactor for the auth layer." },
  ],
});

Try Gemini 3.5 Flash — 30% off list

Same model, same quality, same wire format — at the discounted Anvat effective rate. $2 free credit on signup, no card required.

Get a key →

Related