MiniMax-M3
MiniMax-M3 is an open-weight vision and language model from MiniMax with 427B parameters (mixture of experts). At 4-bit it needs about 283 GB of memory, so the smallest device in our directory that fits it is the Mac Studio M3 Ultra (512GB), which is discontinued.
Overview
Includes room for an 8k context and runtime overhead. More context needs more memory.
Key information
- Lab
- MiniMax
- Released
- 2 Jun 2026
- Parameters
- 427B · MoE
- Type
- Vision and language
- Context
- 1,048,576 tokens
- License
- Custom license
- HF downloads, 30 days
- 137,392
- Repository
- MiniMaxAI/MiniMax-M3
More from MiniMax
View allWill it run on your hardware?
How we estimate| Estimate or device | Memory |
|---|---|
| Q4 | 283 GB |
| Q8 | 499 GB |
| BF16 | 940.6 GB |
| RTX 5060 Ti 16 GB | 16 GB usable |
| RTX 5090 32 GB | 32 GB usable |
| DGX Spark 128 GB | 108.8 GB usable |
| Mac Studio M5 Ultra 256 GB | 192 GB usable |
Single-device memory estimate. Pick a device to update the On your device column.
36 GB usable memory
| Format | Memory needed | On your device | Example hardware |
|---|---|---|---|
| 4-bit (Q4) | ≈ 283 GB | Too big 786% | Mac Studio M3 Ultra (512GB)Discontinued512 GB memory |
| 8-bit (Q8) | ≈ 499 GB | Too big 1386% | No single device in our directory has enough memory |
| 16-bit (BF16) | ≈ 940.6 GB | Too big 2613% | No single device in our directory has enough memory |
Fits means the estimate stays within 90% of usable memory. Tight is above 90% and up to 100%. Too big is above 100%. These are memory checks, not speed or compatibility tests.
Mac allowances are 66% of total RAM up to 36 GB, and 75% above. AI boxes allow 85% because their memory is shared with the operating system.
Where to run it
On a Mac
Start with the MLX files, built for Apple Silicon.
mlx-community/MiniMax-M3-4bitCheck your MacWith GGUF
Choose a file for llama.cpp or a compatible app.
unsloth/MiniMax-M3-GGUFFind compatible hardwareThrough an API
DeepInfra (fp8) is the cheapest provider we found.
$0.224 / $0.88
Input / output per 1M tokens
₹21 / ₹83 at ₹94.29 per $1
Blended: $0.388 (₹37) per 1M. Ranked by a 3:1 mix of input and output tokens.
mlx_vlm.generate --model mlx-community/MiniMax-M3-4bit \
--prompt "Describe this image." --image image.jpgllama-cli -hf unsloth/MiniMax-M3-GGUF \
-c 8192 -p "Explain how a transformer works."Example commands built from the repository ids above. They need a working install of MLX or llama.cpp. We have not run them; check each repository for available files.
| Precision | |||||
|---|---|---|---|---|---|
| DeepInfraCheapest | $0.388₹37 | $0.224₹21 | $0.88₹83 | 524,288 | FP8 |
| CoreWeave | $0.4125₹39 | $0.23₹22 | $0.96₹91 | 262,144 | FP4 |
| GMICloud | $0.42₹40 | $0.24₹23 | $0.96₹91 | 1,048,576 | FP8 |
| Minimax | $0.525₹50 | $0.30₹28 | $1.20₹113 | 524,288 | Not listed |
| Parasail | $0.525₹50 | $0.30₹28 | $1.20₹113 | 1,048,576 | FP8 |
| Venice | $0.525₹50 | $0.30₹28 | $1.20₹113 | 524,288 | FP8 |
INR at ₹94.29 per $1. Prices fetched 14 Nov 2026.
More about this model
Frequently asked questions
How much memory does MiniMax-M3 need? At an 8k context, about 283 GB at 4-bit, 499 GB at 8-bit and 940.6 GB at 16-bit (BF16). A longer context needs more.
Can I run MiniMax-M3 on a Mac? No current Mac fits the 4-bit estimate of 283 GB. The smallest Mac in our directory that does is the Mac Studio M3 Ultra (512GB), which is discontinued. An MLX build is available: mlx-community/MiniMax-M3-4bit.
Does the full context fit? MiniMax-M3 supports up to 1,048,576 tokens of context. Our memory estimates use 8k, so long prompts need more memory than shown.
What does it cost through an API? The cheapest provider we found is DeepInfra, at $0.22 per 1M input tokens and $0.88 per 1M output tokens (₹21 and ₹83 at 94.29 rupees per dollar). Ranked by a 3:1 mix of input and output tokens.
License in plain English
Custom license. This model uses a custom license rather than a standard one. Read it on the model card before using the model commercially. This summary is not legal advice.
OMHQ RigBench
Not yet tested. We show no speed or quality score without a measurement.
Revision history and sources
Last updated . Sources:Hugging Face model card, MLX repository, GGUF repository, OpenRouter.