GEMINI 3.5 LITE GA1M CONTEXTMULTIMODALHIGH THROUGHPUTNO AUDIO SURCHARGE
Gemini 3.5 Flash-Lite API Pricing
Gemini 3.5 Flash-Lite shipped GA on July 21, 2026 for high-volume agentic tasks, translation and simple data processing. It lists $0.30/M input and $2.50/M output, with cache hits at $0.03/M. Note the rate card before switching: per token it is dearer than Gemini 3.1 Flash-Lite. Pulled directly from ai.google.dev daily.
Input - per 1M tokens
$0.30/M
Covers audio too new
Output - per 1M tokens
$2.50/M
Incl. thinking tokens new
Cached input - per 1M tokens
$0.03/M
Storage $1/M-hour -90%
Effective - agentic blend
$0.27/M
92/8 split - 82% cache
§ 01 / TERMINAL
Run the numbers.
Live calculator pre-loaded with current Gemini 3.5 Flash-Lite rates. Tweak spend, output mix, or cache assumptions and share the URL to share the calculation.
$ /mo
Workload split
Prompt cache hit rate
Tokens you can process
—
Words equivalent (English)
—
Effective rate
—
§ 02 / SCENARIOS
Real-world presets.
BULK
High-volume classification
$0.00313/item
ASSISTANT
Research planning turn
$0.00641/turn
CODING AGENT
Codebase-scale migration
$0.034/task
LONG DOC
Reading 100-page contracts
$0.034/doc
§ 03 / TOKENIZER
Paste text. See tokens. See cost.
Estimate · gemini-tokenizer-estimate · ≈3.85 chars/token Auto-counts as you type
This is a chars-per-token approximation, not a real tokenizer. Actual tokens vary by language, code density, and tool-call overhead — counts are typically ±10–20% off for English prose, more for code or non-Latin scripts. For exact billing, use the vendor's official tokenizer.
Characters 683
Words 107
Tokens (estimated) 177 tokens
Cost as input · uncached $0.00005 USD
Cost as output · uncached $0.00044 USD
Cost as cached input $0.00001 USD
| Model | Input /M | Output /M | Effective blended | Context | Best for |
|---|---|---|---|---|---|
| Gemini 3.5 Flash-Lite Current | $0.30 cache $0.03 | $2.50 | $0.27 agentic 92/8 | 1M | High-volume agentic + translation |
| Gemini 3.1 Flash-Lite | $0.25 cache $0.03 | $1.50 | $0.18 cheaper on text | 1M | Cheapest per-token Gemini for text |
| Gemini 2.5 Flash | $0.30 cache $0.03 | $2.50 | $0.27 same rate | 1M | Prior-gen twin of this rate card |
| Gemini 3.6 Flash | $1.50 cache $0.15 | $7.50 | $0.96 pricier | 1M | Step up for harder agentic work |
| GPT-5.4 nano | $0.20 cache $0.02 | $1.25 | $0.15 cheaper | 400K | OpenAI's budget tier |
| DeepSeek V4 Flash | $0.14 cache $0.0028 | $0.28 | $0.05 cheaper | 1M | Budget reasoning and coding |
Frequently asked.
Practical pricing questions, separated from calculator assumptions and regional taxes.
Q · 01 What is Gemini 3.5 Flash-Lite priced at? +
Google's paid-tier pricing page lists
$0.30/M input, $2.50/M output and $0.03/M cached input, plus $1.00 per 1M tokens per hour of explicit cache storage. The single input rate covers text, image, video and audio.Q · 02 Google calls it the most cost-efficient GA model. Is it the cheapest Gemini? +
Not per token. Gemini 3.1 Flash-Lite is cheaper on both sides —
$0.25/M input and $1.50/M output against $0.30 and $2.50 here — and Gemini 2.5 Flash carries an identical rate card. Any saving has to come from this model finishing a task in fewer tokens, which is a claim to test on your own workload, not a lower rate.Q · 03 When is it cheaper than Gemini 3.1 Flash-Lite? +
On audio. 3.1 Flash-Lite charges
$0.50/M for audio input against $0.25/M for text, while this model bills every modality at $0.30/M. Audio-heavy pipelines can therefore land cheaper here; text-only bulk work will not.Q · 04 Does output pricing include thinking tokens? +
Yes. Google's price table labels the output row "including thinking tokens", so reasoning is billed at the same
$2.50/M. Thinking is supported on this model and there is no separate reasoning rate.Q · 05 How is the effective price calculated? +
AI//COST uses the same 92/8 agentic blend everywhere: 92% of tokens are input, 82% of which are served from cache. That gives a blended input of
$0.0786/M and an effective cost of $0.27/M. Every shelf row above uses the same formula.Q · 06 What does Batch or Flex inference cost? +
The model page marks Batch API, Flex and Priority inference as supported, but Google's pricing page publishes no per-model rate for those tiers on this model today. This page records the standard tier only rather than assuming the usual discount.
Q · 07 What context window does it support? +
1,048,576 input tokens and 65,536 output tokens. Inputs cover text, image, video, audio and PDF; output is text only. Computer use, Live API, audio generation and image generation are not supported — for computer use, see Gemini 3.6 Flash.
Q · 08 How accurate is the tokenizer estimate? +
The browser widget uses a
gemini-tokenizer-estimate chars-per-token ratio for English text — Google publishes no downloadable tokenizer, so this is an estimate, not exact counting. Actual billing comes from the API's usage metadata and differs for code, non-Latin scripts and multimodal input.