Estimate Google Gemini 1.5 Pro, Gemini 1.5 Flash, and Gemini 2.0 costs instantly. The cheapest capable LLM API in 2026 — see exactly how much you'd save vs GPT-4o.
Includes savings comparison vs GPT-4o
Google AI Studio / Vertex AI rates in USD per million tokens, June 2026.
| Model | Input / 1M | Output / 1M | Context | Multimodal |
|---|---|---|---|---|
| Gemini 1.5 FlashCHEAPEST | $0.075 | $0.30 | 1M tokens | ✅ Text+Image+Audio |
| Gemini 2.0 Flash | $0.10 | $0.40 | 1M tokens | ✅ Text+Image+Audio |
| Gemini 1.5 Pro | $1.25 | $5.00 | 2M tokens | ✅ Text+Image+Audio+Video |
| Gemini 2.0 Pro | $3.50 | $10.50 | 1M tokens | ✅ Full multimodal |
Google AI Studio offers a free tier — 15 req/min, 1M tokens/day for Gemini 1.5 Flash. Perfect for prototyping before moving to paid production. No credit card required to start.
Gemini 1.5 Flash handles up to 1 million tokens in a single request — entire codebases, long documents, hours of transcripts. At $0.075/M input, analyzing 1M tokens costs $0.075. GPT-4o charges $2.50/M with a 128K limit.
| Workload | GPT-4o | Gemini 1.5 Flash | Savings with Gemini |
|---|---|---|---|
| 1M requests, 1K in + 500 out tokens | $7,500 | $225 | 97% cheaper |
| Long-doc analysis (10K tokens × 10K req) | $25,000 | $750 | 97% cheaper |
| Support chatbot (2K in + 400 out × 50K req) | $45,000 | $1,350 | 97% cheaper |
Gemini 1.5 Flash: $0.075/$0.30 per million tokens — the cheapest capable LLM API. Gemini 2.0 Flash: $0.10/$0.40. Gemini 1.5 Pro: $1.25/$5.00. All June 2026 pay-as-you-go rates via Google AI Studio or Vertex AI.
Yes — dramatically. Gemini 1.5 Flash ($0.075/$0.30) is 33× cheaper than GPT-4o ($2.50/$10.00) on both input and output. Even Gemini 1.5 Pro ($1.25/$5.00) is 2× cheaper than GPT-4o on input. For most high-volume tasks, Gemini is the cheapest option available.
Yes — Google AI Studio provides a free tier with 15 requests/minute and 1M tokens/day for Gemini 1.5 Flash. No credit card required. For production workloads, you'll need a paid Google Cloud account.
For many tasks yes — Gemini 1.5 Flash performs comparably to GPT-4o mini on classification, extraction, summarization, and translation, at half the price. Gemini 1.5 Pro is competitive with GPT-4o on reasoning and coding. For the absolute best reasoning quality, GPT-4o or Claude Sonnet may still be preferred.