Today's price · per 1M tokens
Input
$0.750
per 1M tokens
Output
$3.75
per 1M tokens
Blended
$1.50
blended $/1M — 3:1 weighted input:output
Cached input
$0.075
10% of input — prompt caching
How this price is scoped: Google's Gemini API pricing page no longer has a Gemini 3.7 Flash section (read 2026-10-14). The price shown is the last rate it printed for this model, read 2026-09-16: a dated, two-phase schedule of $0.75 per 1M input tokens through December 31, 2026, rising to $1.50 on January 1, 2027, and $3.75 per 1M output tokens through December 31, 2026, rising to $7.50, with context caching on the same dates ($0.075, then $0.15). A request sent to gemini-3.7-flash today is served by Gemini 3.8 Flash; see that model's page for the rate Google prints now.
Price receipt
We read Google's own pricing page on and found Gemini 3.7 Flash listed with a price on the same card — that read is where the number above comes from. We keep the page text we read; its content hash is 60a63d484d.
Overview
Deprecated: Google replaced it with Gemini 3.8 Flash on October 8, 2026 and routes its API requests there. Google's fast multimodal Flash model, launched 2026-08-13 as the successor to Gemini 3.6 Flash — built for agentic workflows, coding and multi-step reasoning. Last priced at $0.75/1M input and $3.75/1M output on Google's dated schedule. 1M-token context with native image, audio and video input.
Capabilities
struck through = not supportedBenchmark performance
accuracy % · higher is betterSources: Vals AI (GPQA Diamond, high effort) (GPQA Diamond) · Vals AI (SWE-bench Verified, high effort) (SWE-Bench Verified) · Vellum LLM Leaderboard (Terminal-Bench 2.1, AutoBench) · Vals AI (LiveCodeBench v6, high effort) (LiveCodeBench) · Artificial Analysis (text-only, no tools, high) (Humanity's Last Exam) · ARC Prize verified leaderboard, ARC-AGI-2 semi-private eval, High effort (ARC-AGI 2)
- GDPval-AA v2: 1404.1 ELO (Real-world work tasks, human baseline = 1000)
- Arena (arena.ai): 1489.3 ELO (Human preference)
- Arena WebDev: 1588 ELO (Human preference on building web apps)
- Arena Vision: 1293.8 ELO (Human preference on image-input prompts)
- Arena Agent: -0.022 (Score from real agent sessions (IPS); higher is better)
- Arena Agent cost per task: $0.47 (Median measured cost of one real Agent Arena task)
- Epoch Capabilities Index: 158.4 (CI 156.3–160.6) (Epoch AI's index fitted across many benchmarks)
- FrontierMath Tiers 1–3: 71.6% (high effort) (Research-level maths (funded by OpenAI, run by Epoch); best run across the effort settings Epoch tested)
- FrontierMath Tier 4: 36.6% (high effort) (Hardest research maths; best run across the effort settings Epoch tested)
Source: Artificial Analysis (artificialanalysis.ai)
- Ranks #13 of 100 benchmarked models by average score
- Ranks #23 of 189 comparably-measured models by percentile score, across 8 independent measurements
- Ranks #45 of 189 comparably-tested models by normalized performance per dollar
- Strongest at GPQA Diamond — 93.9%, #9 of 71
287.8 tokens/sec output
Throughput: Artificial Analysis, recorded 2026-09-14 · Source: Artificial Analysis (artificialanalysis.ai)
Source: Vellum LLM Leaderboard · updated · See full rankings →
Specifications
- Provider
- Context window
- 1M tokens
- Modality
- text, image, audio, video
- Parameters
- Proprietary
- Open source
- No — proprietary
- Released
- Status
- Deprecated
- Last updated
- Tags
Availability verified: — per Google's own deprecation notice