MiniMax M3 (MiniMax-M3) is a chat model from minimax with a 1M-token context window; open-weight; capabilities tools, vision, reasoning; input image, text, video; output text.
1MContext
128KMax output
$0.3$ / 1M in
18.9Intelligence
106.4tok / s
Pricing
Input
$0.3 / 1M tokens
Output
$1.2 / 1M tokens
Currency
USD
Note
Indicative US list price — verify with the vendor.
Indicative US list price — a reference only, verify with the vendor. Source: litellm. Last verified 2026-07-22.
Benchmarks
Intelligence index
18.9
Cited from a public third-party leaderboard — a reference, verify at the source. Source: Artificial Analysis. Last verified 2026-11-05.
Performance
Throughput
106.4 tokens/s
Latency (TTFT)
0.82 s
Cited speed metrics — a reference, verify at the source. Source: Artificial Analysis. Last verified 2026-11-05.