MinimaxFastReleased 31 May 2026
MiniMax M3 pricing, context and benchmark
MiniMax-M3 is a multimodal foundation model from MiniMax.
1.05M
Input context
524K
Output context
$0.23
Official input
$0.96
Official output
19
AA index
01 · Model Overview
Everything that matters, in one table
Context, pricing, cache, modalities and tooling, from the model registry and Artificial Analysis.
| Model | MiniMax M3 |
| Model ID | minimax/minimax-m3 |
| Provider | Minimax |
| Release date | 31 May 2026 |
| Status | Generally available |
| Input context | 1,048,576 tokens |
| Output context | 524,288 tokens |
| Official input pricing | $0.23 per 1M tokens |
| Official output pricing | $0.96 per 1M tokens |
| Cache read pricing | $0.06 per 1M tokens (80% off input) |
| Market input pricing | $0.224 per 1M (cheapest route) |
| Market output pricing | $0.88 per 1M (cheapest route) |
| Highest route pricing | $0.3 in, $1.2 out per 1M |
| Input modalities | text · image · video |
| Output modalities | text |
| Reasoning control | Not exposed |
| Tool support | Text generation only |
| Tokenizer | Other |
| Provider routes | 8 endpoints |
| Artificial Analysis index | 19 (#16 of 26) |
| Speed | 127 tokens/s, 60.62s to first token |
| Knowledge cutoff | Not publicly stated |
Input and output support
| Modality | Input | Output | Billed at |
|---|---|---|---|
| Text | ✓ yes | ✓ yes | $0.23 in |
| Image | ✓ yes | no | $0.23 in |
| Audio | no | no | not supported |
| Video | ✓ yes | no | $0.23 in |
| File (PDF) | no | no | not supported |
02 · Overview
What is MiniMax M3?
MiniMax-M3 is a multimodal foundation model from MiniMax.
It supports text, image, and video inputs with text output, a 1M-token context window, and is suited for long-horizon agentic work, coding,...
MiniMax M3 carries a 1.05M token window, uses the Other tokenizer and is served by Minimax. On Hoodgen it sits in the Fast tier of the model menu next to its siblings, so you can switch mid conversation without losing the thread.
03 · Release
Release and what changed from the previous version
Released31 May 2026
This is the only release of MiniMax M3 on OpenRouter, released 31 May 2026. The provider has not published a knowledge cutoff, so treat training data as recent but unconfirmed for very new events.
04 · Capabilities
Features and capabilities
◉
Vision
Screenshots, charts and photographed documents are read at the text token rate.
♪
Audio and video
Calls, meetings and screen recordings can be summarised in one pass.
⚡
Prompt caching
Cached input costs $0.06 per million instead of $0.23, so stable prefixes get cheaper.
06 · Benchmarks
Benchmark results and performance
Artificial Analysis Intelligence Index v5.0.1 blends nine evaluations. MiniMax M3 scores 19 , ranking #16 of 26.
Intelligence index
19
#16 of 26
Cost per task
$1.29
weighted average per index task
Time per task
12m
inference time
Output speed
127
tokens per second
Index score and cost per task
Scale: index 0 to 70, cost 0 to $20. Sub scores are not published per evaluation, so none are invented here.
07 · Use cases
What can you do with MiniMax M3?
High volume drafts
Captions, titles, product blurbs and rewrite passes where cost per call dominates.
Inline assistants
Autocomplete, quick replies and menu helpers that must answer inside a second.
Batch classification
Tagging, routing and extraction across thousands of short inputs.
Cheap agent loops
Tool calling loops that do not need frontier grade reasoning.
08 · Cost levers
Prompt caching and how to cut the bill
Cached input bills at $0.06 per million instead of $0.23, a 80% discount. Keep the system prompt and documents first and the changing question last, so every call hits the cache.
09 · Pricing
MiniMax M3 pricing
Official pricing
List price
$0.23 / 1M in
$0.96 / 1M out
- $0.06 cached input
- 1.05M context
- text, image, video input
Cheapest route
$0.224 / 1M in
$0.88 / 1M out
- 8 routes on OpenRouter
- What Hoodgen routes to first
- Same model, lower bill
Highest route
$0.3 / 1M in
$1.2 / 1M out
- Priority or premium tier
- Best under load
- Use for latency critical calls
Market pricing, every route
| Provider | Tier | Input | Output | Cache read | Uptime 24h |
|---|---|---|---|---|---|
| DeepInfra | flex | $0.224 | $0.88 | $0.045 | 100.0% |
| CoreWeave | standard | $0.23 | $0.96 | $0.05 | 99.8% |
| GMICloud | standard | $0.24 | $0.96 | $0.048 | 100.0% |
| DeepInfra | standard | $0.28 | $1.1 | $0.056 | 97.9% |
| StreamLake | standard | $0.3 | $1.2 | $0.06 | 99.8% |
| Venice | standard | $0.3 | $1.2 | $0.06 | 98.1% |
| Parasail | standard | $0.3 | $1.2 | $0.06 | 100.0% |
| Novita | standard | $0.3 | $1.2 | $0.06 | 100.0% |
API pricing and cost calculator
| Rate | Per call | Total | Credits |
|---|---|---|---|
| Cheapest route | $0.0036 | $1.82 | 1734 |
| List price | $0.0040 | $2.00 | 1900 |
| Highest route | $0.0078 | $3.90 | 3705 |
10 · Developers
API and developer information
| Model ID | minimax/minimax-m3 |
| Endpoint | https://openrouter.ai/api/v1/chat/completions |
| Parameters | prompt and max_tokens |
| Streaming | Yes, server sent events |
| Modality string | text+image+video->text |
| Max prompt tokens | 1,048,576 |
| Max completion tokens | 524,288 |
curl https://openrouter.ai/api/v1/chat/completions \
-H "Authorization: Bearer $OPENROUTER_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"model": "minimax/minimax-m3",
"messages": [{ "role": "user", "content": "Summarise this document." }],
"max_tokens": 1024
}'11 · Limitations
Model limitations
| Text only output | It cannot generate images, audio or video. Pair it with an image model on Hoodgen when you need visuals. |
| Output cap | 524,288 tokens per request. Long deliverables have to be chunked. |
| Knowledge cutoff | The provider does not state a fixed cutoff for every release. Verify very recent events against search. |
| Depth ceiling | Fast tier: strong on throughput, weaker than flagship tiers on long multi step reasoning. |
12 · Comparison
MiniMax M3 vs other Minimax models
| Model | Context | Input /1M | Output /1M | AA index | Released |
|---|---|---|---|---|---|
| MiniMax M3 | 1.05M | $0.23 | $0.96 | 19 | 2026 |
13 · Privacy
Privacy and data policy
- On Hoodgen, prompts and attachments answer that request and stay in your conversation history so you can reopen it. They are not used to train models.
- Requests route through OpenRouter to Minimax and are covered by that provider's API terms.
- Do not paste secrets or personal data you would not send to any hosted model.
14 · FAQ
MiniMax M3 questions, answered
MiniMax M3 is a fast model from Minimax, released 31 May 2026. MiniMax-M3 is a multimodal foundation model from MiniMax.
✦
Start chatting with MiniMax M3
Free to try on Hoodgen AI, with a live credit estimate before every send. No API key, no setup.