Skip to content
Metatext Decide — typed decisions for your agents, with a probability on every answerGet early access
Metatext
Sign Up

MiniMax M3 Thinking

NanoGPTOpen weights

MiniMax M3 Thinking by NanoGPT costs $0.30 per 1M input tokens and $1.20 per 1M output tokens, with a 512K-token context window.

Pricing

Input (per 1M tokens)
$0.30
Output (per 1M tokens)
$1.20
Cached input (per 1M)
$0.06

Specifications

Provider
NanoGPT
Context window
512K tokens
Parameters
—
Released
Jun 2026
Open weights
Yes
Frontier model
No

Benchmarks

LiveBench
70.0

Compare MiniMax M3 Thinking with…

MiniMax M3 Thinking vs MiniMax M2.7$1.20/1MMiniMax M3 Thinking vs MiniMax M3$1.20/1MMiniMax M3 Thinking vs GPT-5.6 Luna$1.20/1MMiniMax M3 Thinking vs Qwen3 Next 80B A3B Thinking$1.20/1MMiniMax M3 Thinking vs Inkling Small$1.20/1MMiniMax M3 Thinking vs KAT-Coder-Pro V1$1.20/1M

FAQ

How much does MiniMax M3 Thinking cost?

MiniMax M3 Thinking is priced at $0.30 per 1M input tokens and $1.20 per 1M output tokens. Output tokens cost more than input.

What is MiniMax M3 Thinking's context window?

MiniMax M3 Thinking supports a context window of 512K tokens — the maximum prompt plus response length it can handle in a single request.

Is MiniMax M3 Thinking open weight?

Yes — MiniMax M3 Thinking has openly available weights.

Data as of 2026-10-16. Source: models.dev (open). Pricing is per 1M tokens (USD); confirm with the provider before production use.