NeuroGenesysGemini 3.8 Flash API pricing
Last updated • By Will Schulz
Gemini 3.8 Flash costs $1.50 per 1M input tokens and $7.50 per 1M output tokens on Google's standard API tier, as of Nov 09, 2026. An introductory rate of $0.75 input and $3.75 output applies until Dec 31, 2026. 1,000 chatbot replies (2K tokens in, 500 out each) cost about $6.75.
How much does Gemini 3.8 Flash cost per token?
| Rate | Input | Output | Cached input |
|---|---|---|---|
| Standard | $1.50 | $7.50 | not listed |
| Introductory, until Dec 31, 2026 | $0.75 | $3.75 | not listed |
Standard-tier list prices from Google's pricing page, last checked Nov 09, 2026. Batch processing and committed-use discounts can be lower; cache-write surcharges and taxes are not included.
What do common jobs cost on Gemini 3.8 Flash?
| Workload | Cost | At intro price | |
|---|---|---|---|
| 1,000 chatbot replies2K tokens in, 500 out each | $6.75 | $3.38 | calculator |
| Summarize 1M tokens of documents100 documents of 10K tokens, a 500-token summary each | $1.88 | $0.94 | calculator |
| 1,000 RAG queries20K tokens in (half from the prompt cache), 800 out eachno cache price listed: billed at the full input rate | $36.00 | $18.00 | calculator |
| 100 coding-agent steps150K tokens in (90% cached), 4K out eachno cache price listed: billed at the full input rate | $25.50 | $12.75 | calculator |
| 1M classification calls800 tokens in, 10 out each | $1,275 | $638 | calculator |
Computed with the LLM API cost calculator's pricing rules: cached tokens at the cache-read rate where one is listed, and the long-context rate for the whole request once a prompt passes the threshold. Tokenizers differ, so the same text is a different number of tokens on each provider.
How does Gemini 3.8 Flash compare with similar-priced models?
| Model | Input / 1M | Output / 1M | 1,000 chatbot replies | vs Gemini 3.8 Flash |
|---|---|---|---|---|
| Claude Haiku 4.5Anthropic | $1.00 | $5.00 | $4.50 | 0.67× |
| GPT-6.1 TerraOpenAI | $1.00 | $5.00 | $4.50 | 0.67× |
| Gemini 3.8 Flash | $1.50 | $7.50 | $6.75 | — |
| Claude Sonnet 5.5Anthropic | $2.00 | $10.00 | $9.00 | 1.33× |
| GPT-6.1 SolOpenAI | $2.00 | $10.00 | $9.00 | 1.33× |
| Gemini 3.1 Pro (preview)Google | $2.00 | $12.00 | $10.00 | 1.48× |
Closeness and order use the list-price cost of 1,000 chatbot replies (2K tokens in, 500 out each). Compare every model on your workload.
Questions
How much does Gemini 3.8 Flash cost per 1M tokens?
$1.50 per 1M input tokens and $7.50 per 1M output tokens at Google's standard list price, as of Nov 09, 2026, with an introductory $0.75 / $3.75 until Dec 31, 2026. Output tokens include any reasoning tokens the model generates.
How much do 1,000 chatbot replies cost with Gemini 3.8 Flash?
About $6.75 at the list price, for 1,000 replies of 2,000 input and 500 output tokens each ($3.38 at the introductory rate). Longer prompts, prompt caching and reasoning tokens change this; the cost calculator handles all three.
Does Gemini 3.8 Flash have prompt caching or long-context pricing?
Our price table lists no cached-input rate for Gemini 3.8 Flash, so cached tokens are priced at the full input rate here. There is no separate long-context rate in the price table.
What is cheaper than Gemini 3.8 Flash?
Among the models priced closest to it, Claude Haiku 4.5 (Anthropic, $1.00 in / $5.00 out, $4.50 per 1,000 chatbot replies) and GPT-6.1 Terra (OpenAI, $1.00 in / $5.00 out, $4.50 per 1,000 chatbot replies) cost less. Price is not quality: compare them on your own prompts before switching.
Related tools
- Free toolLLM API cost calculatorCost per call, day and month across Claude, GPT and Gemini, with caching and long-context rates.
- Free toolToken counterCount a text's tokens and check which context windows it fits.
- GuideLLM API pricing: every modelInput and output prices for every model we track, side by side.
Sources
- Google API pricing — Google
