MiniMax M3 API pricing
Updated
Among the 10 providers we checked (prices read 2026-11-14 to 2026-11-15), the lowest blended standard price for MiniMax M3 is CoreWeave Forge at USD 0.23 per 1M input tokens and USD 0.96 per 1M output tokens. The median across 10 standard prices is USD 0.30 / USD 1.20. MiniMax's own API charges USD 0.30 / USD 1.20.
What does MiniMax charge for MiniMax M3?
- Input
- USD 0.30
- Cached input
- USD 0.06
- Output
- USD 1.20
- Blended 3:1
- USD 0.525
USD per 1M tokens on MiniMax, model MiniMax-M3, standard tier. Source, checked 2026-11-14.
Other MiniMax tiers: priority USD 0.45 / USD 1.80.
Is a host cheaper than MiniMax? Yes. CoreWeave Forge charges USD 0.23 input and USD 0.96 output per 1M tokens, 21% below MiniMax's own USD 0.30 / USD 1.20 on the 3:1 blend.
MiniMax M3 model facts
Facts from MiniMax's model card for MiniMaxAI/MiniMax-M3 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-M3) with no access gate. | Hugging Face model API, checked 2026-11-16Quote“"private": false, "gated": false, 59 .safetensors files listed” |
| Licence | Custom licence: minimax-community | MiniMax model card metadata, checked 2026-11-16Quote“license: other; license_name: minimax-community” |
| Total parameters | ~428B | MiniMax model card, checked 2026-11-16Quote“It has ~428B parameters and ~23B activated parameters.” |
| Active parameters per token | ~23B | MiniMax model card, checked 2026-11-16Quote“It has ~428B parameters and ~23B activated parameters.” |
| Context length | 1M | MiniMax model card, checked 2026-11-16Quote“MiniMax-M3 is a native multimodal model with 1M context.” |
| Input and output | Image and text in, text out (the card's task tag is image-text-to-text) | MiniMax model card metadata, checked 2026-11-16Quote“pipeline_tag: image-text-to-text” |
| Tool calling | Not found on MiniMax's model card when we read it | checked 2026-11-16 |
| First published on Hugging Face (not necessarily the release date) | 2026-06-02 | Hugging Face model API, checked 2026-11-16Quote“"createdAt": "2026-06-02T07:49:31.000Z"” |
MiniMax M3 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 2.3× the cheapest. Free rows, the maker's serving version and other service tiers follow, tagged and unranked.
| # | Provider | Input | Cached input | Output | Blended 3:1 | Context | Precision | Source |
|---|---|---|---|---|---|---|---|---|
| 1 | CoreWeave Forge cheapestMiniMaxAI/MiniMax-M3Notes and source quote“MiniMax M3 $0.23 $0.96 $0.05” | USD 0.23 | USD 0.05 | USD 0.96 | USD 0.4125 | 262,144 | not stated | CoreWeave Forge page checked 2026-11-15 re-checked 2026-11-16 |
| 2 | GMI Cloud promoMiniMaxAI/MiniMax-M3Notes and source quotePromotional price, 20% off list (list 0.3 in / 1.2 out / 0.06 cached). Long-prompt tier: prompts over 512,000 tokens cost 0.48 in / 1.92 out. “Price 20% off ≤ 512k token Input $0.24 $0.30 Output $0.96 $1.20 Cache Read $0.048 $0.06 Cache Write - > 512k token Input $0.48 $0.60 Output $1.92 $2.40” | USD 0.24 | USD 0.048 | USD 0.96 | USD 0.42 | 1,048,576 | FP8 | GMI Cloud page checked 2026-11-15 re-checked 2026-11-16 |
| 3 | DeepInfraMiniMaxAI/MiniMax-M3Notes and source quote“$0.28 in $1.10 out $0.056 cached / 1M tokens” | USD 0.28 | USD 0.056 | USD 1.10 | USD 0.485 | 524,288 | FP8 | DeepInfra page checked 2026-11-14 re-checked 2026-11-16 |
| 4 | MiniMax makerMiniMax-M3Notes and source quotePrompts up to 512K input tokens, after the "Permanent 50% off" MiniMax shows on its list price (0.6 input, 2.4 output, 0.12 cache read). Prompts over 512K cost double. Cached price is the prompt-caching read price. “| **MiniMax-M3** ≤ 512k input tokens Permanent 50% off | ~~$0.60~~ $0.30 | ~~$2.40~~ $1.20 | ~~$0.12~~ $0.06 |” | USD 0.30 | USD 0.06 | USD 1.20 | USD 0.525 | not stated | not stated | MiniMax page checked 2026-11-14 re-checked 2026-11-16 |
| 5 | Fireworks AIaccounts/fireworks/models/minimax-m3Notes and source quote“MiniMax M3 | $0.30 / $0.06 / $1.20 | $0.45 / $0.09 / $1.80” | USD 0.30 | USD 0.06 | USD 1.20 | USD 0.525 | 512,000 | not stated | Fireworks AI page checked 2026-11-14 re-checked 2026-11-16 |
| 6 | Hyperbolicminimax-m3Notes and source quoteContext shown on page as "1M" (rounded); stored as the decimal value. “MiniMax M3 minimax-m3 MiniMax · 1M context · Text $0.30 $1.20 $0.06” | USD 0.30 | USD 0.06 | USD 1.20 | USD 0.525 | 1,000,000 | not stated | Hyperbolic page checked 2026-11-14 re-checked 2026-11-16 |
| 7 | ParasailNotes and source quote“MiniMax M3 $0.30 $1.20 $0.06” | USD 0.30 | USD 0.06 | USD 1.20 | USD 0.525 | not stated | not stated | Parasail page checked 2026-11-14 re-checked 2026-11-16 |
| 8 | SiliconFlowNotes and source quoteContext shown as 1049K. “MiniMax-M3 1049K $0.3 $0.06 $1.2” | USD 0.30 | USD 0.06 | USD 1.20 | USD 0.525 | 1,048,576 | not stated | SiliconFlow page checked 2026-11-14 re-checked 2026-11-16 |
| 9 | Microsoft AzureNotes and source quoteFoundry Models serverless, served by Fireworks on Azure, Data Zone deployment. "MiniMax 3" mapped to MiniMax M3. “FW MiniMax 3 Data Zone Input: $0.33/1M Tokens | Cached Input: $0.066/1M Tokens | Output: $1.32/1M Tokens” | USD 0.33 | USD 0.066 | USD 1.32 | USD 0.5775 | not stated | not stated | Microsoft Azure page checked 2026-11-14 re-checked 2026-11-16 |
| 10 | FeatherlessMiniMaxAI/MiniMax-M3Notes and source quoteDeveloper plan (per-token billing, from 50 USD a month in credits). “PRICING: INPUT $0.55 / CACHED $0.06 / OUTPUT $2.2” | USD 0.55 | USD 0.06 | USD 2.20 | USD 0.9625 | 262,144 | FP8 | Featherless page checked 2026-11-15 re-checked 2026-11-16 |
| — | DeepInfra flex / spotMiniMaxAI/MiniMax-M3Notes and source quoteFlex service tier (0.8× standard). “Flex (0.8×) $0.224 $0.88 $0.0448” | USD 0.224 | USD 0.0448 | USD 0.88 | USD 0.388 | 524,288 | FP8 | DeepInfra page checked 2026-11-14 re-checked 2026-11-16 |
| — | MiniMax maker priorityMiniMax-M3Notes and source quotePriority tier (service_tier=priority, 1.5x standard). Prompts up to 512K input tokens, after the "Permanent 50% off" MiniMax shows on its list price (0.9 input, 3.6 output, 0.18 cache read). Prompts over 512K cost double. Cached price is the prompt-caching read price. “| **MiniMax-M3** ≤ 512k input tokens Permanent 50% off | ~~$0.90~~ $0.45 | ~~$3.60~~ $1.80 | ~~$0.18~~ $0.09 |” | USD 0.45 | USD 0.09 | USD 1.80 | USD 0.7875 | not stated | not stated | MiniMax page checked 2026-11-14 re-checked 2026-11-16 |
| — | Fireworks AI priorityaccounts/fireworks/models/minimax-m3Notes and source quotePriority mode. “MiniMax M3 | Priority $0.45 / $0.09 / $1.80” | USD 0.45 | USD 0.09 | USD 1.80 | USD 0.7875 | 512,000 | not stated | Fireworks AI 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.
- Promo: the provider's page describes this price as promotional or discounted (GMI Cloud). It ranks as published, but can end without notice; the row's notes give the list price or end date where the page states one.
- Other tiers (flex / spot and priority): batch runs asynchronously within hours; flex and spot are best-effort and can be slow or refused at capacity; priority and fast cost more for reliability or speed. Shown, not ranked.
- 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 M3 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) |
|---|---|---|---|
| CoreWeave Forge | USD 28.26 | USD 56.64 | USD 225.00 |
| GMI Cloud (promo) | USD 28.51 | USD 58.18 | USD 224.64 |
| DeepInfra | USD 32.90 | USD 67.57 | USD 260.28 |
| MiniMax (maker) | USD 35.64 | USD 72.72 | USD 280.80 |
| Fireworks AI | USD 35.64 | USD 72.72 | USD 280.80 |
| Hyperbolic | USD 35.64 | USD 72.72 | USD 280.80 |
| Parasail | USD 35.64 | USD 72.72 | USD 280.80 |
| SiliconFlow | USD 35.64 | USD 72.72 | USD 280.80 |
| Microsoft Azure | USD 39.20 | USD 79.99 | USD 308.88 |
| Featherless | USD 62.64 | USD 129.72 | USD 438.30 |
For the support chatbot, the cheapest price (CoreWeave Forge, USD 28.26 a month) saves USD 7.38 against the median (USD 35.64). For the coding agent the gap is USD 56.16 a month (20% lower). The winner changes with the workload: GMI Cloud wins the cache-heavy agent, CoreWeave Forge the chatbot.
Which MiniMax M3 provider is cheapest for your traffic?
CoreWeave Forge has the lowest input price (USD 0.23) and the lowest output price (USD 0.96), 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 |
|---|---|---|---|
| CoreWeave Forge | USD 0.4125 cheapest | USD 0.345 cheapest | USD 0.291 |
| GMI Cloud | USD 0.42 | USD 0.348 | USD 0.2904 cheapest |
| DeepInfra | USD 0.485 | USD 0.401 | USD 0.3338 |
| MiniMax (maker) | USD 0.525 | USD 0.435 | USD 0.363 |
| Fireworks AI | USD 0.525 | USD 0.435 | USD 0.363 |
| Hyperbolic | USD 0.525 | USD 0.435 | USD 0.363 |
USD per 1M tokens, 3 input tokens to 1 output token, each provider's cheapest standard row; 4 dearer providers are in the price table above. 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.
All 10 providers publish a cached-input price below the input price: CoreWeave Forge (78% off), GMI Cloud (80% off), DeepInfra (80% off), MiniMax (80% off), Fireworks AI (80% off) and 5 more.
Context windows differ: 1,048,576 tokens at GMI Cloud and SiliconFlow; 1,000,000 tokens at Hyperbolic; 524,288 tokens at DeepInfra; 512,000 tokens at Fireworks AI; 1 more sizes in the table. MiniMax, Parasail and Microsoft Azure do not state one. The cheapest row is not always the one that fits a long prompt.
What does MiniMax M3 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 M3:
| Content | Input per 1M characters | Cheapest provider |
|---|---|---|
| English | USD 0.048 | CoreWeave Forge |
| Arabic MSA | USD 0.071 | CoreWeave Forge |
| Hindi | USD 0.137 | CoreWeave Forge |
| Chinese | USD 0.128 | CoreWeave Forge |
| Code | USD 0.059 | CoreWeave Forge |
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.
Published benchmark results
BroadNet did not run these benchmarks; results are as published by Epoch AI and may use different settings. BroadRouter is BroadNet's site.
| Benchmark | Result | Setting or note | Source |
|---|---|---|---|
| GPQA Diamond | 0.909 | Setting as listed by Epoch (no suffix); Epoch run MiniMax-M3, mean score on a 0 to 1 scale | Epoch AI, checked 2026-11-16 |
Epoch AI rows: Epoch AI, Capabilities & benchmarking, licensed CC BY 4.0, shown as the mean score on a 0 to 1 scale, rounded to three decimals (hover a value for the exact figure in Epoch's data file).
Estimate your own bill
- LLM API cost calculator with MiniMax M3 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 M3 compare with other MiniMax models?
The MiniMax models whose cheapest standard price is closest to MiniMax M3's (USD 0.4125 blended):
| Model | Cheapest input | Cheapest output | Max context | Providers |
|---|---|---|---|---|
| MiniMax M2.5 | USD 0.27 | USD 0.95 | 204,800 | 6 |
| MiniMax M2.7 | USD 0.21 | USD 0.84 | 204,800 | 3 |
| MiniMax M2 | USD 0.2775 | USD 1.11 | 204,800 | 4 |
| Minimax M2.1 | USD 0.30 | USD 1.20 | 204,800 | 3 |
| MiniMax M3.1 Flash | USD 0.15 | USD 0.60 | 1,048,576 | 9 |
Priced closest from other makers: Qwen3 Next 80B A3B Thinking (Qwen, USD 0.4125) · MiMo V2.6 Pro (Xiaomi, USD 0.4078) · DeepSeek V3.1 (DeepSeek, USD 0.425).
Compare MiniMax M3 with
- MiniMax M3 vs MiniMax M2.5: API price side by side
- GLM 5.3 vs MiniMax M3: API price side by side
- Kimi K3 vs MiniMax M3: API price side by side
- Qwen3.8 27B vs MiniMax M3: API price side by side
- DeepSeek V4.1 Flash vs MiniMax M3: API price side by side
- MiniMax M3 vs MiniMax M2.7: API price side by side
Also compared: vs MiniMax M2.5 · vs GLM 5.3 · vs Kimi K3 · vs Qwen3.8 27B · vs DeepSeek V4.1 Flash · vs MiniMax M2.7 · vs MiniMax M3.1 Flash.
Every MiniMax M3 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 M3 API cost per token?
From USD 0.23 per 1M input tokens (USD 0.00000023 per token) and USD 0.96 per 1M output tokens (USD 0.00000096 per token) at the lowest standard prices we found. The median of 10 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 M3?
Among the 10 providers we checked (prices read 2026-11-14 to 2026-11-15), the lowest is CoreWeave Forge, ranked on a blend of 3 input tokens to 1 output token (USD 0.4125 per 1M). For input alone the lowest price is USD 0.23 (CoreWeave Forge); for output alone, USD 0.96 (CoreWeave Forge and 1 other, tied). Batch, flex, priority and fast tiers are shown in the table but not ranked.
Is MiniMax M3 cheaper from a host than from MiniMax?
Yes. CoreWeave Forge charges USD 0.23 input and USD 0.96 output per 1M tokens, 21% below MiniMax's own USD 0.30 / USD 1.20 on the 3:1 blend.
Does MiniMax M3 have cached input pricing?
All 10 standard prices include one. The lowest is USD 0.048 per 1M at GMI Cloud; the median is USD 0.06. Providers without one bill cached input at the input price. What cached input means.
What is the MiniMax M3 context window?
Up to 1,048,576 tokens. 5 standard prices come with less, down to 262,144 tokens (CoreWeave Forge, Featherless, Fireworks AI, DeepInfra and Hyperbolic). 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 M3?
For 1,500 requests a day, 40,000 input tokens (85% cached), 2,000 output incl. reasoning, over 30 days: USD 224.64 a month at GMI Cloud, USD 280.80 at the median price and USD 438.30 at Featherless. Try your own numbers.
Is there a batch, flex or priority price for MiniMax M3?
Yes: flex / spot from DeepInfra at USD 0.224 / USD 0.88; priority from MiniMax at USD 0.45 / USD 1.80 per 1M input / output tokens. These tiers trade speed or availability for price (batch runs within hours, flex and spot can be slow or refused at capacity, priority and fast cost more for speed or reliability), so they are shown but not ranked.
Is MiniMax M3 free to use through an API?
No provider in our data lists a free price for MiniMax M3 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 M3 open source, with open weights, and what is its licence?
MiniMax publishes the weights on Hugging Face as MiniMaxAI/MiniMax-M3, in a public repository with no access gate. The licence, as the MiniMax model card metadata states it: Custom licence: minimax-community. 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 M3 have?
Total parameters, as the MiniMax model card states them: ~428B. Active parameters per token: ~23B. (checked 2026-11-16; BroadNet did not verify these figures.)
How do I call MiniMax M3 with the OpenAI SDK?
BroadNet does not serve MiniMax M3. Use the OpenAI-compatible base URL of a provider that lists it: create the client with OpenAI(base_url=..., api_key=...) and pass that provider's model ID. Documented base URLs for providers on this page, in alphabetical order: DeepInfra (https://api.deepinfra.com/v1/openai, model MiniMaxAI/MiniMax-M3) and Fireworks AI (https://api.fireworks.ai/inference/v1, model accounts/fireworks/models/minimax-m3). Each provider page quotes the provider's docs with a check date.
How does MiniMax M3 score on benchmarks?
As published, Epoch AI measured 0.909 on GPQA Diamond in its own run, as a mean score on a 0 to 1 scale (checked 2026-11-16). BroadNet did not run these benchmarks, and sources may use different settings, so the figures are not directly comparable. Every published result we found, with its setting and source.
Read as MarkdownAsk ChatGPTAsk ClaudeAsk Perplexity
Sources
- Microsoft Azure: pricing (checked 2026-11-14)
- CoreWeave Forge: pricing (checked 2026-11-15)
- DeepInfra: pricing (checked 2026-11-14)
- Featherless: pricing (checked 2026-11-15)
- Fireworks AI: pricing (checked 2026-11-14)
- GMI Cloud: pricing (checked 2026-11-15)
- Hyperbolic: pricing (checked 2026-11-14)
- MiniMax: own API pricing (checked 2026-11-14)
- Parasail: pricing (checked 2026-11-14)
- SiliconFlow: pricing (checked 2026-11-14)