MiniMax M2.5 API pricing
Updated
Among the 6 providers we checked (prices read 2026-11-14 to 2026-11-15), the lowest blended standard price for MiniMax M2.5 is Venice at USD 0.27 per 1M input tokens and USD 0.95 per 1M output tokens. The median across 6 standard prices is USD 0.30 / USD 1.20. Each price is read from the provider's own pricing page.
Does MiniMax sell MiniMax M2.5 directly?
MiniMax publishes no price for MiniMax M2.5 on its own API that we could find, so this page compares hosts only.
MiniMax M2.5 model facts
Facts from MiniMax's model card for MiniMaxAI/MiniMax-M2.5 and the Hugging Face model API, read 2026-11-16. BroadNet did not verify them, and hosts may serve the model at a different precision. BroadRouter is BroadNet's site.
| Fact | As stated | Source |
|---|---|---|
| Open weights | Yes. The weights are in a public Hugging Face repository (MiniMaxAI/MiniMax-M2.5) with no access gate. | Hugging Face model API, checked 2026-11-16Quote“"private": false, "gated": false, 125 .safetensors files listed” |
| Licence | Custom licence: modified-mit | MiniMax model card metadata, checked 2026-11-16Quote“license: other; license_name: modified-mit” |
| Parameters in the weight files (Hugging Face count, not the maker's figure) | About 228.7 billion (228,689,764,864) | Hugging Face model API, checked 2026-11-16Quote“"safetensors": { "total": 228689764864 }” |
| Context length | Not found on MiniMax's model card when we read it; see the context column in the price table | checked 2026-11-16 |
| Input and output | Text in, text out (the card's task tag is text-generation) | MiniMax model card metadata, checked 2026-11-16Quote“pipeline_tag: text-generation” |
| Tool calling | Yes | MiniMax model card, checked 2026-11-16Quote“thanks to improvements in capabilities such as parallel tool calling” |
| First published on Hugging Face (not necessarily the release date) | 2026-02-12 | Hugging Face model API, checked 2026-11-16Quote“"createdAt": "2026-02-12T06:05:24.000Z"” |
MiniMax M2.5 price per provider
USD per 1M tokens. Standard prices are ranked on a blend of 3 input tokens to 1 output token, cheapest first; the dearest standard price is 19% above the cheapest.
| # | Provider | Input | Cached input | Output | Blended 3:1 | Context | Precision | Source |
|---|---|---|---|---|---|---|---|---|
| 1 | Venice cheapestminimax-m25Notes and source quote“| MiniMax M2.5 | `minimax-m25` | $0.27 | $0.95 | $0.03 | — | Private |” | USD 0.27 | USD 0.03 | USD 0.95 | USD 0.44 | 198,000 | not stated | Venice page checked 2026-11-15 re-checked 2026-11-16 |
| 2 | FeatherlessMiniMaxAI/MiniMax-M2.5Notes and source quoteDeveloper plan (per-token billing, from 50 USD a month in credits). “PRICING: INPUT $0.295 / CACHED $0.059 / OUTPUT $1.2” | USD 0.295 | USD 0.059 | USD 1.20 | USD 0.5212 | 204,800 | FP8 | Featherless page checked 2026-11-15 re-checked 2026-11-16 |
| 3 | Amazon BedrockNotes and source quoteTable headed "Regions: US East (N. Virginia), US East (Ohio) and US West (Oregon)". “MiniMax M2.5 $ 0.30 $ 1.20” | USD 0.30 | not published | USD 1.20 | USD 0.525 | not stated | not stated | Amazon Bedrock page checked 2026-11-14 re-checked 2026-11-16 |
| 4 | DigitalOceanNotes and source quoteDigitalOcean-hosted; public preview. “MiniMax M2.5 (Public Preview) — Processing modePrompt lengthInputOutputCache read: Standard All prompts$0.30$1.20$0.06” | USD 0.30 | USD 0.06 | USD 1.20 | USD 0.525 | not stated | not stated | DigitalOcean page checked 2026-11-14 re-checked 2026-11-16 |
| 5 | GMI CloudMiniMaxAI/MiniMax-M2.5Notes and source quote“Price (per 1M tokens) Input $0.30 Output $1.20 Cache Read $0.03 Cache Write -” | USD 0.30 | USD 0.03 | USD 1.20 | USD 0.525 | 196,608 | FP8 | GMI Cloud page checked 2026-11-15 re-checked 2026-11-16 |
| 6 | ParasailNotes and source quote“MiniMax M2.5 $0.30 $1.20 $0.03” | USD 0.30 | USD 0.03 | USD 1.20 | USD 0.525 | not stated | not stated | Parasail page checked 2026-11-14 re-checked 2026-11-16 |
- Ranked: standard-tier prices only, on (3 × input + output) ÷ 4. "Cached input: not published" means the provider's page shows no cache price; budget cached tokens at the input price.
- Automated re-check 2026-11-16: a weekly job re-fetches each source page and looks for the recorded quote or prices. It is not a human review. "Re-checked" means it still found them. Rows with no re-check note show only the date the price was read.
- Precision is shown only where the provider states it; "not stated" is unknown, not full precision.
Report a wrong price or number (email to info@broadnet.ai, with this page's link filled in).
What does MiniMax M2.5 cost for a real workload?
Monthly cost over 30 days at each standard price, using the presets of the LLM API cost calculator. Providers without a cached price bill cached input at the input price.
- Support chatbot: 2,000 requests a day, 1,500 input tokens (60% cached), 300 output.
- RAG over documents: 1,000 requests a day, 8,000 input tokens (30% cached), 500 output.
- Coding agent (one developer): 1,500 requests a day, 40,000 input tokens (85% cached), 2,000 output incl. reasoning.
| Provider | Support chatbot | RAG over documents | Coding agent (one developer) |
|---|---|---|---|
| Venice | USD 28.44 | USD 61.77 | USD 204.30 |
| Featherless | USD 35.41 | USD 71.81 | USD 277.92 |
| Amazon Bedrock | USD 48.60 | USD 90.00 | USD 648.00 |
| DigitalOcean | USD 35.64 | USD 72.72 | USD 280.80 |
| GMI Cloud | USD 34.02 | USD 70.56 | USD 234.90 |
| Parasail | USD 34.02 | USD 70.56 | USD 234.90 |
For the support chatbot, the cheapest price (Venice, USD 28.44 a month) saves USD 6.27 against the median (USD 34.71). For the coding agent the gap is USD 52.11 a month (20% lower).
Which MiniMax M2.5 provider is cheapest for your traffic?
Venice has the lowest input price (USD 0.27) and the lowest output price (USD 0.95), so it is the cheapest standard price at every mix of input and output tokens, before cache discounts.
| Provider | No cache hits | 50% cached | 90% cached |
|---|---|---|---|
| Venice | USD 0.44 cheapest | USD 0.35 cheapest | USD 0.278 cheapest |
| Featherless | USD 0.5212 | USD 0.4327 | USD 0.3619 |
| Amazon Bedrock (no cache price) | USD 0.525 | USD 0.525 | USD 0.525 |
| DigitalOcean | USD 0.525 | USD 0.435 | USD 0.363 |
| GMI Cloud | USD 0.525 | USD 0.4237 | USD 0.3427 |
| Parasail | USD 0.525 | USD 0.4237 | USD 0.3427 |
USD per 1M tokens, 3 input tokens to 1 output token, each provider's cheapest standard row. A provider without a cached price bills cached input at the input price. Any charge for writing to the cache, which some providers add, is not included.
5 of 6 providers publish a cached-input price below the input price: Venice (89% off), Featherless (80% off), DigitalOcean (80% off), GMI Cloud (90% off), Parasail (90% off). Amazon Bedrock publishes none, so cached tokens there bill at the input price.
Context windows differ: 204,800 tokens at Featherless; 198,000 tokens at Venice; 196,608 tokens at GMI Cloud. Amazon Bedrock, DigitalOcean and Parasail do not state one. The cheapest row is not always the one that fits a long prompt.
What does MiniMax M2.5 cost in other languages?
The same text needs a different number of tokens in each language. With the MiniMax M3 tokenizer, sending 1M characters costs this much at the cheapest standard input price we found for MiniMax M2.5:
| Content | Input per 1M characters | Cheapest provider |
|---|---|---|
| English | USD 0.057 | Venice |
| Arabic MSA | USD 0.084 | Venice |
| Hindi | USD 0.160 | Venice |
| Chinese | USD 0.151 | Venice |
| Code | USD 0.070 | Venice |
From BroadRouter's real cost by language dataset, which covers 15 languages and content types for every model we could match to a measured tokenizer.
Estimate your own bill
- LLM API cost calculator with MiniMax M2.5 preloaded: your traffic and cache rate, against any of 428 models.
- MiniMax token counter: count the tokens in your real prompts. Tokenizers differ, so the same text costs a different number of tokens on different models.
- Prompt caching calculator: what a higher cache-hit rate saves.
How does MiniMax M2.5 compare with other MiniMax models?
The MiniMax models whose cheapest standard price is closest to MiniMax M2.5's (USD 0.44 blended):
| Model | Cheapest input | Cheapest output | Max context | Providers |
|---|---|---|---|---|
| MiniMax M3 | USD 0.23 | USD 0.96 | 1,048,576 | 10 |
| MiniMax M2 | USD 0.2775 | USD 1.11 | 204,800 | 4 |
| Minimax M2.1 | USD 0.30 | USD 1.20 | 204,800 | 3 |
| MiniMax M2.7 | USD 0.21 | USD 0.84 | 204,800 | 3 |
| MiniMax M3.1 Flash | USD 0.15 | USD 0.60 | 1,048,576 | 9 |
Priced closest from other makers: DeepSeek V3 (DeepSeek, USD 0.4375) · Qwen3 Coder 480B A35B (Qwen, USD 0.4375) · Step 3.7 Flash (StepFun, USD 0.4375).
Compare MiniMax M2.5 with
- MiniMax M3 vs MiniMax M2.5: API price side by side
- MiniMax M2.5 vs Minimax M2.1: API price side by side
- MiniMax M2.7 vs MiniMax M2.5: API price side by side
Also compared: vs MiniMax M3 · vs Minimax M2.1 · vs MiniMax M2.7.
Every MiniMax M2.5 price comparison · All model-vs-model price comparisons
All 428 models: API prices compared · Cheapest LLM APIs right now · LLM API price index (open data) · Download all prices (CSV, JSON) · Price change log
Questions
How much does the MiniMax M2.5 API cost per token?
From USD 0.27 per 1M input tokens (USD 0.00000027 per token) and USD 0.95 per 1M output tokens (USD 0.00000095 per token) at the lowest standard prices we found. The median of 6 standard prices is USD 0.30 input and USD 1.20 output. Each price was read from the provider's own pricing page; the source is linked in its row.
Which provider is cheapest for MiniMax M2.5?
Among the 6 providers we checked (prices read 2026-11-14 to 2026-11-15), the lowest is Venice, ranked on a blend of 3 input tokens to 1 output token (USD 0.44 per 1M). It also has the lowest input and the lowest output price. Batch, flex, priority and fast tiers are shown in the table but not ranked.
Does MiniMax sell MiniMax M2.5 on its own API?
MiniMax publishes no price for MiniMax M2.5 on its own API that we could find, so there is no maker price to compare. The lowest-priced host we checked is Venice at USD 0.27 / USD 0.95 per 1M tokens.
Does MiniMax M2.5 have cached input pricing?
5 of 6 standard prices include one. The lowest is USD 0.03 per 1M at Venice; the median is USD 0.03. Providers without one bill cached input at the input price. What cached input means.
What is the MiniMax M2.5 context window?
Up to 204,800 tokens. 2 standard prices come with less, down to 196,608 tokens (GMI Cloud and Venice). Check the context column before you pick the cheapest row for long prompts. Context window calculator.
How much does a coding agent cost per month on MiniMax M2.5?
For 1,500 requests a day, 40,000 input tokens (85% cached), 2,000 output incl. reasoning, over 30 days: USD 204.30 a month at Venice, USD 256.41 at the median price and USD 648.00 at Amazon Bedrock. Try your own numbers.
Is MiniMax M2.5 free to use through an API?
No provider in our data lists a free price for MiniMax M2.5 as of 2026-11-15. Free tiers and credits by provider (checked 2026-11-14) lists general offers that do not name a model; all free LLM API models are on one page.
Is MiniMax M2.5 open source, with open weights, and what is its licence?
MiniMax publishes the weights on Hugging Face as MiniMaxAI/MiniMax-M2.5, in a public repository with no access gate. The licence, as the MiniMax model card metadata states it: Custom licence: modified-mit. Open weights means you can download and run the model yourself; what you may do with it, including commercial use, is set by that licence, so read it in full (checked 2026-11-16). These facts come from MiniMax's card and the Hugging Face model API; BroadNet did not verify them, and hosts may serve the model at a different precision.
How many parameters does MiniMax M2.5 have?
MiniMax's model card does not state a total that we found. Hugging Face counts about 228.7 billion (228,689,764,864) parameters in the published weight files of MiniMaxAI/MiniMax-M2.5. That is a file-level count, not the maker's figure, and it can include extra modules (checked 2026-11-16).
Does MiniMax M2.5 support tool calling?
Yes, according to the MiniMax model card: “thanks to improvements in capabilities such as parallel tool calling” (checked 2026-11-16). Whether the tools parameter works through an API also depends on the provider serving it, so check that provider's docs. BroadNet did not test it.