MiniMax M2.5 API Pricing
Shuts down November 18, 2026. MiniMax dropped MiniMax-M2.5 from its pricing page around the launch of MiniMax M3.1 Flash on October 19, 2026, and has since published a schedule: closed to new users from October 19 and retired on November 18, 2026, when API requests stop working. Its last published rates were $0.3/M input, $1.2/M output and $0.03/M cached input.
MiniMax M2.5 API cost calculator.
Calculator pre-loaded with the last rates MiniMax published for MiniMax M2.5. Tweak workload split and cache hit rate, then share the URL to share the calculation.
MiniMax M2.5 cost on real workloads.
Repo repair task
Analyst memo draft
Web research brief
Product support turn
Counts use a chars-per-token calibration measured on the vendor's own published tokenizer (MiniMaxAI/MiniMax-M2.1, 2026-06-10). English prose is typically within a few percent; code and non-Latin scripts tokenize heavier. For billing-exact counts use the vendor's count-tokens API.
| Model | Input /M | Output /M | Effective blended | Context | Best for |
|---|---|---|---|---|---|
| MiniMax M2.7 | $0.30 cache $0.06 | $1.20 | $0.191 deprecated · +13% | 205K | MiniMax flagship agent loops |
| MiniMax M2.7 Highspeed | $0.60 cache $0.06 | $2.40 | $0.337 deprecated · +2.0× | 205K | Faster MiniMax flagship loops |
| MiniMax M2.5 Current | $0.30 cache $0.03 | $1.20 | $0.168 agentic 92/8 | 205K | Value-first MiniMax agent loops |
| GPT-6.1 Terra | $1.00 cache $0.05 | $5.00 | $0.645 +3.8× | 1.05M | OpenAI's fast, efficient GPT-6 tier |
| DeepSeek V4 Pro | $1.32 cache $0.044 | $3.96 | $0.569 +3.4× | 1M | Low-cost reasoning workloads |
| Gemini 3.8 Flash | $0.75 cache $0.075 | $3.75 | $0.481 +2.9× · promo | 1M | Global multimodal workloads |
MiniMax M2.5 has published scores on 2 benchmarks we track; its best placing is #59 of 99 on ARC-AGI-1. All of these scores come from independent leaderboards, not from the lab.
| Benchmark | Score | Rank | Leader | Source |
|---|---|---|---|---|
| ARC-AGI-1 Reasoning | 63.7% ARC Prize verified, best result | #59 of 99 | 6 tied · 98.5% | Official leaderboard ARC Prize Foundation |
| ARC-AGI-2 Reasoning | 4.9% ARC Prize verified, best result | #71 of 105 | GPT-6 Astra · 95% | Official leaderboard ARC Prize Foundation |
MiniMax M2.5 API pricing FAQ.
Practical MiniMax M2.5 pricing questions, with live MiniMax list rates separated from workload assumptions.
Q · 01 Can I still use MiniMax M2.5? +
MiniMax-M2.5 no longer work. M Plan and Token Plan requests are routed to MiniMax M3.1 Flash from October 19. The rates on this page are the last ones MiniMax published.Q · 02 What is the standard MiniMax M2.5 API price? +
MiniMax-M2.5 at $0.3/M input and $1.2/M output. Cache reads are listed at $0.03/M. AI//COST stores those direct USD list prices without currency conversion.Q · 03 Does MiniMax publish prompt caching for this model? +
$0.03/M and cache writes at $0.375/M for MiniMax M2.5. The quote board uses the read price because repeated cache hits drive recurring workload cost.Q · 04 What context window does MiniMax M2.5 support? +
204,800 tokens of context for this M2-family text model. This page rounds that to 205K for display consistency.Q · 05 Is there a Batch API discount? +
MiniMax M2.5. Treat the public $0.3/M input and $1.2/M output rates as the default unless your account has a negotiated enterprise agreement.Q · 06 When was this price last checked? +
Q · 07 How accurate is the tokenizer estimate? +
minimax-tokenizer-estimate chars-per-token approximation for English planning. Real billing depends on MiniMax server-side tokenization and can differ for Chinese, code, and mixed-language prompts.