Google: Gemini 3.1 Pro Preview
Gemini 3.1 Pro Preview is a reasoning model from Google released on Feb 18, 2026. It reads text, images, files, audio and video and writes text, with a 1.05M token context window and up to 66K tokens per reply.
Show moreShow less
It costs $2 per million input tokens and $12 per million output tokens. Cached input bills at $0.20, 90% less than fresh input.
It supports tool calling, structured JSON output and prompt caching.
31 providers serve it, including Vercel AI Gateway, Vertex and Snowflake Cortex. The lowest input price is $1 per million tokens at Kilo Gateway.
On the Artificial Analysis Intelligence Index it scores 29.7, which places it at number 23 among the 43 models I track with independent scores.
- Modalities
- In / out price
- $2 / $12 per 1M
- Context
- 1.05M
- Max output
- 66K
- Released
- Feb 18, 2026
- Knowledge cutoff
- Jan 2025
- Tool calling
- Structured output
- Reasoning
- Prompt caching
- Open weights
- Free to use
Gemini 3.1 Pro Preview pricing
USD, as listed on Oct 15, 2026. Most bills are input heavy, so the cached input rate matters as much as the headline price.
| Input | $2 / 1M tokens | |
|---|---|---|
| Output | $12 / 1M tokens | |
| Cached input (read) | $0.20 / 1M tokens | 90% off input |
| Prompts over 200K | $4 in, $18 out | long context rate |
| Prompts over 200K | $4 in, $18 out | long context rate |
Gemini 3.1 Pro Preview providers
31 providers serve it, including Vercel AI Gateway, Vertex and Snowflake Cortex. The lowest input price is $1 per million tokens at Kilo Gateway. Each host sets its own price and limits, so the cheapest one is not always the one with the longest context.
| Provider | Input /M | Output /M | Cache read /M | Context | Max output |
|---|---|---|---|---|---|
| Google First party | $2 | $12 | $0.20 | 1.05M | 66K |
| Vercel AI Gateway | $2 | $12 | $0.20 | 1M | 64K |
| Vertex | $2 | $12 | $0.20 | 1.05M | 66K |
| Snowflake Cortex | n/a | n/a | – | 1.05M | 66K |
| Kilo Gateway | $1 | $6 | $0.10 | 1.05M | 66K |
| 302.AI | $2 | $12 | – | 1.05M | 66K |
| Abacus | $2 | $12 | $0.20 | 1.05M | 66K |
| AIHubMix | $2 | $12 | $0.20 | 1.05M | 66K |
| Auriko | $2 | $12 | $0.20 | 1.05M | 66K |
| CrossModel | $2 | $12 | $0.20 | 1.05M | 66K |
| DaoXE | $2 | $12 | $0.20 | 1.05M | 66K |
| DevPass (LLM Gateway) | $2 | $12 | $0.20 | 1.05M | 66K |
Showing 12 of 31 providers: the lab first, then the major clouds, then the cheapest resellers.
What Gemini 3.1 Pro Preview costs per month
Pick a workload close to yours or type your own volumes.
Independent benchmarks
Measured by third parties, not reported by Google. Each links to its source and shows the date I checked it.
A composite of 10 independent evaluations (agents 30%, coding 20%, scientific reasoning 20%, general 30%), run by Artificial Analysis on each model, version 4.3.2.
Artificial Analysis, checked Oct 10, 2026
Human preference: people compare two anonymous answers and pick the better one. Ratings from the Text Arena overall board, style control on.
Arena (formerly LMArena), checked Oct 10, 2026
Abstract reasoning on visual puzzles the model has not seen, semi private set. Hard to memorise, so it tracks reasoning more than knowledge.
ARC Prize, checked Oct 10, 2026
Median output tokens per second measured by Artificial Analysis on the provider API, for the same model setting as its index score.
Artificial Analysis, checked Oct 10, 2026
Benchmarks Google reports (17)
- hle-no-tools (Humanity's Last Exam, full set, text + MM, no tools, Gemini 3.1 Pro Thinking (High))44.4%
- hle-with-tools (Humanity's Last Exam, full set, Search (blocklist) + Code)51.4%
- arc-agi-2 (ARC-AGI-2, ARC Prize Verified)77.1%
- gpqa-diamond (GPQA Diamond, no tools)94.3%
- terminal-bench-2 (Terminal-Bench 2.0, Terminus-2 harness)68.5%
- swe-bench-verified (SWE-Bench Verified, single attempt)80.6%
- swe-bench-pro (SWE-Bench Pro (Public), single attempt)54.2%
- livecodebench-pro (LiveCodeBench Pro, competitive coding (Codeforces, ICPC, IOI))2887
- scicode (SciCode, scientific research coding)59%
- apex-agents (APEX-Agents, long horizon professional tasks)33.5%
- gdpval-aa (GDPval-AA Elo, expert tasks (v1, not GDPVal-AA v2))1317
- tau2-bench (τ2-bench, Retail)90.8%
- mcp-atlas (MCP Atlas, multi-step workflows using MCP)69.2%
- browsecomp (BrowseComp, Search + Python + Browse)85.9%
- mmmu-pro (MMMU-Pro, no tools)80.5%
- mmmlu (MMMLU, multilingual Q&A)92.6%
- mrcr-v2 (MRCR v2 (8-needle), 128k (average))84.9%
Labs pick their own settings and harnesses, so compare these only within one lab.
Supported request parameters
What you can set when you call Gemini 3.1 Pro Preview through an OpenAI compatible API.
- Tools
tools - Structured outputs
structured_outputs - Reasoning
reasoning - Temperature
temperature
Alternatives to Gemini 3.1 Pro Preview
Models that take the same inputs and call tools, closest in price, cheaper ones first.
- 1.05M context
Reads text, images, files, audio and video. Tool calling, structured output, reasoning. Served by 30 providers.
by googleFeb 5, 2025$1.25/M input tokens$10/M output tokens30 providers
- 1.05M context
Reads text, images, files, audio and video. Tool calling, structured output, reasoning. Served by 12 providers.
by googleFeb 19, 2026$2/M input tokens$12/M output tokens12 providers
- 1.05M context
Reads text, images, files, audio and video. Tool calling, structured output, reasoning.
by vivgridFeb 19, 2026$2/M input tokens$12/M output tokens
- 1.05M context
Reads text, images, files, audio and video. Tool calling, structured output, reasoning. Served by 7 providers.
by googleNov 18, 2025$2/M input tokens$12/M output tokens7 providers
- 1.05M context
Reads text, images, files, audio and video. Tool calling, structured output, reasoning. Served by 13 providers.
by googleSep 30, 2026$2/M input tokens$10/M output tokens13 providers
- 1.05M context
Reads text, images, files, audio and video. Tool calling, structured output, reasoning. Served by 32 providers.
by googleMay 19, 2026$1.50/M input tokens$9/M output tokens32 providers
More from Google
Nano Banana 2.1
Image66K contextGenerates images from text, images and files. Served by 4 providers.
by googleOct 6, 2026$1.50/M input tokens$7.50/M output tokens4 providers
- 1.05M context
Reads text, images, files, audio and video. Tool calling, structured output, reasoning. Served by 13 providers.
by googleSep 30, 2026$2/M input tokens$10/M output tokens13 providers
Turns text into spoken audio.
by googleSep 23, 2026$0.50/M input tokens$6/M output tokens
Gemini 3.8 Flash TTS
SpeechTurns text into spoken audio.
by googleSep 23, 2026$0.50/M input tokens$9/M output tokens
- 262K context
Reads text. Reasoning.
by googleSep 19, 2026$0.05/M input tokens$0.15/M output tokens
- 262K context
Reads text and images. Tool calling, reasoning.
by googleSep 19, 2026$0.1056/M input tokens$0.3344/M output tokens
Gemini 3.1 Pro Preview questions
How much does the Gemini 3.1 Pro Preview API cost?
$2 per million input tokens and $12 per million output tokens, with cached input at $0.20. Prompts over 200K tokens bill at $4 input and $18 output. The estimator on this page turns that into a monthly bill for your own workload.
What is the context window of Gemini 3.1 Pro Preview?
1,048,576 tokens of input, and it can write up to 65,536 tokens in one reply. As a rough guide, a million tokens is around 750,000 English words.
Which providers serve Gemini 3.1 Pro Preview?
31 providers serve it, including Vercel AI Gateway, Vertex and Snowflake Cortex. The lowest input price is $1 per million tokens at Kilo Gateway. Prices differ by host, so the Providers table on this page lists each one with its input and output price.
Does Gemini 3.1 Pro Preview support tool calling?
Yes. It accepts tool definitions and can return JSON that matches a schema, which is what an agent needs to call your APIs.
Is Gemini 3.1 Pro Preview open source?
No. Google serves it through an API only, so you cannot download the weights or host it yourself.
When was Gemini 3.1 Pro Preview released?
Feb 18, 2026, with training data up to Jan 2025.
How good is Gemini 3.1 Pro Preview compared with other models?
It scores 29.7 on the Artificial Analysis Intelligence Index and 1487 on the Arena text leaderboard. Benchmarks are a shortlist, not a verdict: test the top two or three candidates on your own tasks before you commit.
Catalog data from models.dev (MIT licensed), refreshed daily, last on Oct 15, 2026. Independent scores from my hand checked rankings, sources linked above. The summary is written from these facts.
Building on Gemini 3.1 Pro Preview?
I test your real tasks on it and the two closest alternatives, then ship the agent on whichever wins on cost and accuracy.
Book Meeting