What Gemini 3 Pro Thinking costs for real workloads
| Workload | Estimated cost |
|---|---|
| Support chatbot: 100K messages | $230.00 |
| RAG search: 10K queries with 6K-token context | $85.00 |
| Batch summarization: 1,000 long documents | $25.00 |
| Coding agent: 100 tasks, 50K context x 20 turns | $19.75 |
Chatbot: 800 input + 300 output tokens per message. RAG: 6,000 + 500 per query. Summarization: 20,000 + 1,000 per document. Coding agent: 20 turns per task over a 50K-token context, with 95% of input served from cache where the model supports it.
Gemini 3 Pro Thinking vs similarly priced models
| Model | Provider | Input / 1M | Output / 1M | Cached / 1M | Blended |
|---|---|---|---|---|---|
| Gemini 2.0 Live | $0.75 | $4.50 | — | $1.69 | |
| DeepSeek V4 Pro 0813 | Together AI | $1.32 | $3.96 | $0.13 | $1.98 |
| gpt-6.1-terra | OpenAI | $1.00 | $5.00 | $0.050 | $2.00 |
| Gemini 3 Pro Lite | $1.00 | $5.00 | $0.050 | $2.00 | |
| Gemini 3 Pro Flash | $1.00 | $5.00 | $0.050 | $2.00 | |
| Gemini 3 Pro Thinking | $1.00 | $5.00 | $0.050 | $2.00 | |
| Z.ai GLM 5.3 | Mistral AI | $1.40 | $4.40 | $0.14 | $2.15 |
| GLM-5.3 | Together AI | $1.40 | $4.40 | $0.26 | $2.15 |
| GLM-5.2 | Together AI | $1.40 | $4.40 | $0.26 | $2.15 |
Other Google models
Gemini 3 Pro Thinking pricing FAQ
How much does the Gemini 3 Pro Thinking API cost?
Gemini 3 Pro Thinking costs $1.00 per million input tokens and $5.00 per million output tokens on Google, with cached input at $0.050 per million. Prices as of November 8, 2026.
How much does Gemini 3 Pro Thinking cost for 1,000 requests?
At 1,000 input and 500 output tokens per request, 1,000 requests cost about $3.50.
Does Gemini 3 Pro Thinking have a prompt caching discount?
Yes. Cached input is $0.050 per million tokens, 20x cheaper than fresh input. Workloads that resend the same context, like agents and chat with long system prompts, pay mostly the cached rate.
Is Gemini 3 Pro Thinking expensive compared to other LLM APIs?
It ranks #48 cheapest of 76 models we track on a blended price (3 input tokens per output token) of $2.00 per million tokens.
Picking a model for production?
Describe your workload and the Planner estimates monthly token spend across every model we track.