Skip to content

MiniMax M3

ModelOpen sourceActive
by MiniMax · family “minimax-m”

MiniMax-M3 is a multimodal foundation model from MiniMax. It supports text, image, and video inputs with text output, a 1M-token context window, and is suited for long-horizon agentic work, coding,... (description from the OpenRouter listing)

in: textin: imagein: videoout: textReasoningTool useStructured output
Epoch Capabilities Index
147.0 #62 of 268
90% CI 142.9 – 149.7
Listing ↗
Released
1 Jun 2026
Context window
1.05M tokens
Max output
512K tokens
Input $ / 1M tokens
$0.30
Output $ / 1M tokens
$1.20
Cached input $ / 1M
$0.060
Each benchmark shown separately with its own source

Benchmark results (12)

BenchmarkDomainScorevs best recordedSettingRunSource
LMCAagents33.7%
49%
——External ↗
DTBenchreasoning78.9%
80%
——External ↗
Mystery Game Puzzlesgames8.0% ±2.7
10%
none27 Aug 2026Epoch ↗
OSWorld 2.0agents4.6%
15%
——External ↗
Surface Evolver Benchscience55.0%
58%
——External ↗
FrontierCodecoding14.7%
27%
——External ↗
ProofBenchmath18.0%
18%
——External ↗
APEX-Agentsagents37.7%
50%
——External ↗
Chess Puzzlesgames14.0% ±3.5
19%
—10 Aug 2026Eval log ↗
OTIS Mock AIME 2024-2025math71.1% ±6.8
71%
—10 Aug 2026Eval log ↗
SimpleBenchreasoning45.8%
56%
——External ↗
GPQA diamondscience90.9% ±2.0
95%
—10 Aug 2026Eval log ↗

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

5 recorded prices on 6 Jun 2026 (OpenRouter listing + Internet Archive snapshots); re-read on every data refresh, most recently 10h ago. Steps show when the price changed.

Events