Skip to content

Gemini 3.1 Pro Preview

ModelActive
by Google · family “gemini-pro-preview”

Gemini 3.1 Pro Preview is Google’s frontier reasoning model, delivering enhanced software engineering performance, improved agentic reliability, and more efficient token usage across complex workflows. Building on the multimodal foundation... (description from the OpenRouter listing)

in: audioin: filein: imagein: textin: videoout: textReasoningTool useStructured output
Epoch Capabilities Index
Not scored by Epoch AI
Listing ↗
Released
19 Feb 2026
Context window
1.05M tokens
Max output
66K tokens
Input $ / 1M tokens
$2.00
Output $ / 1M tokens
$12.0
Cached input $ / 1M
$0.20
Each benchmark shown separately with its own source

Benchmark results (32)

BenchmarkDomainScorevs best recordedSettingRunSource
Furniture Assemblymultimodal26.7% ±5.6
32%
high10 Sep 2026Epoch ↗
LMCAagents53.8%
79%
high—External ↗
DTBenchreasoning97.1%
98%
medium—External ↗
Mystery Game Puzzlesgames34.0% ±4.8
40%
high27 Jul 2026Epoch ↗
EBR-benchreasoning14.3%
19%
—25 Jun 2026Epoch ↗
MirrorCodecoding8.9% ±4.6
12%
high10 Aug 2026Epoch ↗
FrontierMath-Tiers-1-3-v2-Privatemath59.6% ±2.9
64%
—11 Jun 2026Eval log ↗
FrontierMath-Tier-4-v2-Privatemath26.8% ±7.0
27%
—11 Jun 2026Eval log ↗
DeepSWEcoding11.7%
16%
——External ↗
ExploitBenchcoding26.1%
35%
——External ↗
CL-bench Lifelong-context16.9%
76%
——External ↗
PostTrainBenchagents22.0%
53%
——External ↗
CL-benchlong-context20.8%
75%
——External ↗
ProofBenchmath26.0%
26%
——External ↗
APEX-Agentsagents35.3%
47%
——External ↗
Chess Puzzlesgames55.0% ±5.0
76%
—19 Feb 2026Eval log ↗
SimpleQA Verifiedknowledge73.5% ±1.4
97%
high10 Aug 2026Epoch ↗
FrontierMath-Tier-4-2025-07-01-Privatesupersededmath16.7% ±5.4
35%
—19 Feb 2026Epoch ↗
DeepResearch Benchagents47.8%
86%
high—External ↗
GSO-Benchcoding22.6%
48%
——External ↗
Terminal Benchagents80.2% ±2.6
95%
agent: TongAgents19 Feb 2026External ↗
ARC-AGI-2reasoning77.1%
81%
——External ↗
METR Time Horizonsagents77.0%
90%
——External ↗
FrontierMath-2025-02-28-Privatesupersededmath36.9% ±2.8
70%
—19 Feb 2026Epoch ↗
HLEknowledge46.4%
85%
——External ↗
WeirdMLcoding72.1% ±0.0
77%
——External ↗
OTIS Mock AIME 2024-2025math95.6% ±3.1
96%
—20 Feb 2026Eval log ↗
Balroggames57.0%
83%
——External ↗
SimpleBenchreasoning79.6%
97%
——External ↗
SWE-Bench verifiedcoding75.6% ±2.0
91%
—24 Feb 2026Eval log ↗
GPQA diamondscience94.4% ±1.6
99%
high6 Aug 2026Eval log ↗
ARC-AGIreasoning98.0%
99%
——External ↗

Source: Epoch AI Benchmarking Hub (CC BY 4.0). “External” rows are leaderboard results Epoch collects from third parties. Best reported setting per benchmark is shown.

API list price over time

Price history

$ per 1M tokens

8 recorded prices on 3 Mar 2026 (OpenRouter listing + Internet Archive snapshots); re-read on every data refresh, most recently 7h ago. Steps show when the price changed.

Events