Get alerts
All products/LLM APIs/Gemini API

Gemini API pricing, price hikes and cheaper alternatives

Gemini API (ai.google.dev) is a LLM API. Pay per token, with input and output priced separately per 1M tokens (thinking tokens bill at the output rate), and some Pro models charging more above 200K-token prompts.. Free plan: Free of charge on most models, but content is used to improve Google's products; Gemini 4 Argon and 3.1 Pro Preview have no free tier..

What Gemini API costs

  • Top model: $40/mo (Gemini 4 Argon (gemini-4-argon). The LMArena leaderboard returned HTTP 403, so its rank is unverified; the pick is the newest-generation Gemini model and also the most expensive non-Pro text model, which is the flagship fallback. Cost = 10 x $2.00 input + 2 x $10.00 output = $40.00 a month. From Dec 1, 2026 the price is $4.00 / $20.00, which would make it $80.00.)
  • Cheapest model: $5.50/mo (Gemini 3.1 Flash-Lite (gemini-3.1-flash-lite) at $0.25 text input and $1.50 output per 1M tokens. Cost = 10 x 0.25 + 2 x 1.50 = $5.50 a month. Gemini 2.5 Flash-Lite ($0.10 / $0.40 = $1.80 a month) would be cheaper, but access to the 2.5 models was limited to past users on 2026-09-18, so it is excluded. The pricing page states no context window or tool-calling support for 3.1 Flash-Lite; both are assumed from the Gemini family.)
  • gpt-oss-120b: not offered (n/a: gpt-oss-120b is not sold per token on the Gemini API, so the point is not offered.)

Plan compared: Pay-as-you-go Standard paid tier, text only, list prices as of 2026-10-17. Monthly USD cost of one reference workload: 10,000 requests a month, totaling 10M input and 2M output tokens, at standard pay-as-you-go rates with no caching or batch discounts. Top model is the product's highest-ranked model on the LMArena Text leaderboard. Cheapest model is its lowest-cost model with tool calling and a 128K context. gpt-oss-120b compares hosts on the same open model.

What Gemini API does

Core: Frontier-grade models, 1M-token context, Ongoing free tier.

Important: Image and PDF input, Voice and realtime audio, Image and video generation, Hosted agent tools, Prompt caching discounts, Batch and flex pricing.

Also has: Open-weight models (partly), OpenAI-compatible endpoint (partly), Ultra-fast inference (partly), Embeddings and reranking (partly), Flat-rate coding plan (partly), No-training, zero retention (partly), Compliance and data residency (partly).

Gemini API price hikes

Alternatives to Gemini API

Gemini API is the cheapest LLM API we track at 12M tokens a month on its top model. The others:

Gemini API alternatives: 3 LLM APIs compared, at every size.

Other kinds of products that cover part of Gemini API

  • Hugging Face (GPU cloud & model hosting): covers 43% of Gemini API; missing or partial: Frontier-grade models (partly), 1M-token context (partly), Ongoing free tier, Image and PDF input (partly), Voice and realtime audio (partly)
  • Perplexity (AI assistants): covers 41% of Gemini API; missing or partial: 1M-token context (partly), Ongoing free tier, Image and PDF input, Voice and realtime audio, Image and video generation

Every change we logged

  • Sep 18, 2026 · Limits cut: Access to the Gemini 2.5 models was restricted to users who have actively used them in the past. The models are not deprecated and remain served until further notice; no price change was stated.

Sources

Pricing page: ai.google.dev/gemini-api/docs/pricing. Prices last checked Oct 16, 2026. Prices come from the official Gemini API pricing page (Standard paid tier) and the vendor changelog, both checked 2026-10-17. The LMArena leaderboard returned HTTP 403 and the web search budget for this run was used up, so the top-model pick is unverified; announced increases (Gemini 4 Argon to $4 / $20 on 2026-12-01, and Gemini 3.6 and 3.8 Flash doubling on 2027-01-01) are kept out of changes until they take effect.