GoogleTextVisionAudio

gemini-3.1-pro

gemini-3.1-pro is Google's model at $2.00 input and $12.00 output per million tokens, with a 1049K context window. On a typical 3:1 input-to-output mix that blends to roughly $4.50 per million tokens, placing it in the premium band of the 89 models tracked here — cheaper than 16% of them.

Open cost calculator View provider Open official platform

MODEL OVERVIEW

Context window1049K

Input price$2 / 1M

Output price$12 / 1M

Output speed70 tok/s

Quality score91

Data updatedUpdated Jun 2026

Price verified2026-08-19 · Official

By price, gemini-3.1-pro ranks 61st cheapest of 89 on input and 74th on output. Its quality score of 91 ranks 23rd of 89, and on value (quality score ÷ output price) it ranks 65th of 89. Its blended rate sits 46% above the catalog average.

In real billing terms: a support chatbot handling 50,000 turns a month (2,000 input / 500 output tokens each) costs about $500 per month on gemini-3.1-pro, while a RAG workload of 20,000 queries (8,000 input / 800 output tokens each) runs about $512. Because output is priced 6.0× input here, long-generation tasks scale cost fastest — cap output length before you optimise anything else.

A 1049K context window holds roughly 1,399 pages of A4 text (at ~750 tokens per page), ranking 8th of 89 here. That is enough to drop in an entire technical manual or several source files at once, which suits long-document summarization and codebase-wide analysis. Output runs at about 70 tok/s (61st fastest of 89), acceptable interactively, though long responses will feel slow.

We track 8 Google models, and gemini-3.1-pro's blended rate sits 109% above that provider's average. If your task tolerates a little less headroom, qwen3.5 (Alibaba) blends about 98% cheaper at a quality score of 91, and is worth benchmarking first. Going the other way, deepseek-reasoner-v4 scores 92 on quality for about 88% less.

The closest model in Google's line-up is gemini-2.5-pro ($1.25 input / $10.00 output, quality 93). Against it, gemini-3.1-pro scores 2 points lower on quality, runs 5 tok/s faster (about 8%), costs 24% more blended. When two models sit this close on price, the deciding factor is usually the quality ceiling — run both against your real prompts before picking.

Our read on it

Google's Pro tier, sitting at a price that is squeezed from three directions at once.

The positioning problem behind $2/$12 and quality 91

gemini-3.1-pro brings a 1,049K context, 70 tokens/second, text, vision and audio, and a quality score of 91. It is a complete spec sheet. The difficulty is what the $12 output rate has to be measured against.

claude-sonnet-5 is $2/$10 at quality 92, with the same 1,049K context and 95 tokens/second. Now that Anthropic has cancelled the September increase, it matches gemini-3.1-pro on input and undercuts it by 17% on output, while carrying one more point of quality and 36% more throughput — the column has flipped entirely. The harder comparison is still grok-4.5 at $2/$6 and quality 91: the same input rate, half the output rate, the same score. Read purely as numbers, gemini-3.1-pro struggles to win a column against grok-4.5.

Its real moat does not appear in a price table

People choose it for position rather than rate. If your data already lives in Google Cloud, auth runs through GCP IAM, BigQuery is the source of record and Vertex AI is the deployment path you already operate, then moving to another vendor carries egress charges, a fresh security review and a second monitoring stack. None of that shows up in a dollars-per-million-tokens column, and for enterprise buyers it is regularly decisive.

By the same logic, price gemini-2.5-pro alongside it: $1.25/$10 at quality 93 is both cheaper and higher-scoring within Google's own lineup, with a much smaller context window. If you are staying in the ecosystem regardless, the choice between those two mostly reduces to whether you need a million tokens.

70 tokens/second decides which layer it belongs in

Seventy tokens/second is mid-pack and rated Medium latency. That suits analytical, asynchronous work: question answering across large document sets, multimodal report generation, research tasks that need to read everything before answering.

It does not suit a token-by-token user interface. If your product is a chat window, Google's own gemini-3.1-flash-lite at $0.25/$1.50 or gpt-5.6-luna from OpenAI will feel dramatically better and cost an order of magnitude less. The sensible architecture puts a fast model on the front line and calls 3.1-pro only when a task genuinely needs reasoning across a million tokens.

Estimated monthly cost

Monthly API spend across four common workloads, next to the cheapest model we track.

Workload	Usage assumption	This model	Cheapest model
Support chatbot	2,000 input / 500 output tokens per turn, 50,000 turns per month	$500	$3.00
RAG document Q&A	8,000 input / 800 output tokens per query, 20,000 queries per month	$512	$3.84
Coding assistant	12,000 input / 2,000 output tokens per request, 10,000 requests per month	$480	$3.20
Batch summarization	4,000 input / 400 output tokens per item, 200,000 items per month	$2,560	$19

Derived directly from the rates on this page. Excludes prompt caching, batch discounts and free tiers.

How it compares

The alternatives worth benchmarking alongside it, from the same dataset.

Model	Provider	Blended /1M	Quality	Price delta
gemini-3.1-pro	Google	$4.50	91	—
qwen3.5Cheaper option	Alibaba	$0.11	91	-98%
deepseek-reasoner-v4Higher quality	DeepSeek	$0.54	92	-88%
gemini-2.5-proSame provider	Google	$3.44	93	-24%
qwen3.7-flashBest value overall	Alibaba	$0.055	82	-99%

Best for

CodingReasoningLong ContextMultimodal

Modalities

TextVisionAudio

Where it fits — and where it doesn't

Good fit

Coding: writing, refactoring and debugging across multi-file project context.
Reasoning: multi-step decomposition, planning and tasks that need rigorous inference.
Long documents: a 1049K window ingests a full contract, paper or large codebase in one pass.
Multimodal input: handles images, screenshots and scanned documents directly.

Poor fit

High-frequency batch jobs: at $4.50 blended per million tokens, bills scale quickly at volume.

Frequently asked questions

How much does the gemini-3.1-pro API cost?

gemini-3.1-pro costs $2.00 per million input tokens and $12.00 per million output tokens. On a 3:1 input-to-output mix that blends to about $4.50 per million tokens.

What does gemini-3.1-pro cost per month?

It depends on volume. Using the four worked scenarios on this page: a support chatbot runs about $500/month, RAG document Q&A about $512/month, a coding assistant about $480/month, and batch summarization about $2,560/month. Plug your own token counts into the cost calculator for a tailored figure.

How large is the gemini-3.1-pro context window?

1049K tokens — roughly 1,399 pages of A4 text at ~750 tokens per page, ranking 8th of the 89 models tracked here.

gemini-3.1-pro or qwen3.5 — which is better value?

qwen3.5 blends to about $0.11 per million tokens versus $4.50 for gemini-3.1-pro — roughly 98% cheaper — at a quality score of 91 against 91. If that quality gap does not show up on your task, take the cheaper one; if it does, stay with gemini-3.1-pro. Benchmark both on your own data before deciding.

What is gemini-3.1-pro best suited for?

Tagged use cases: coding, reasoning, long-context, multimodal. Coding: writing, refactoring and debugging across multi-file project context. Less suitable: High-frequency batch jobs: at $4.50 blended per million tokens, bills scale quickly at volume.

How current is this gemini-3.1-pro pricing?

Pricing here was last verified on 2026-08-19 against official documentation. llmprice.app runs a daily collection job, but providers can change rates between runs — confirm against Google's official pricing page before you commit.