About
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,...
Related Models
Pricing
Input
$0.30
per 1M tokens
Output
$1.20
per 1M tokens
Blended
$0.52
per 1M tokens
Cheaper than 51% of models. Median price is $0.54/1M tokens.
Cost Calculator
Tokens per day1M
100K100M
Daily
$0.52
Monthly
$15.75
vs. Similar Models
Gemini 3.5 Flash (medium)Q:+0.1
$3.38+543%
Google: Gemini 3.5 FlashQ:+0.6
$3.38+543%
Kimi K2.6Q:-0.8
$1.36+160%
Xiaomi: MiMo-V2.5-ProQ:-0.9
$0.54+4%
Performance
44
tokens/sec
Faster than 10% of models
2.16
seconds
Faster than 24% of models
48.03
seconds
Faster than 8% of models
Market Median
93 tok/s
53% slower
Median TTFT
1.09s
98% slower
Throughput/Dollar
83
tok/s per $/1M
Speed Comparison
Claude 4 Opus (Reasoning)
44 tok/s+0%
Llama 3 Instruct 70B
43 tok/s-0%
QwQ 32B-Preview
43 tok/s-1%
Context Window
1.0M
tokens
Larger than 90% of models
Max Output
512K
tokens
49% of context
Benchmarks
MMLU-ProNot evaluated
GPQA Diamond
92.9%
HLE
37.1%
LiveCodeBenchNot evaluated
SciCode
45.4%
TerminalBench Hard
42.4%
MATH-500Not evaluated
AIMENot evaluated
AIME 2025Not evaluated
IFBench
82.9%
Long Context Recall
74.0%
Tau2
88.9%
Market AverageTop Score