MiniMax: MiniMax M3.
MiniMax-M3 is a native multimodal mixture-of-experts model, about 428B parameters with 23B active and a 1M-token context, with adjustable reasoning, under a community licence.
2× AMD MI300X · $1.00/hr.
Most-aggressive quantisation we have a working recommendation for. Lower precision = less VRAM = cheaper hardware, at a small accuracy cost.
Cheapest hosted endpoints.
Variants in the MiniMax family.
MiniMax: MiniMax M3.1 Flash
MiniMax-M3.1-Flash is a multimodal model from MiniMax. Compared with M3, it d...
MiniMax: MiniMax M1
MiniMax-M1 is a large-scale, open-weight reasoning model designed for extende...
MiniMax: MiniMax M2
MiniMax-M2 is a compact, high-efficiency large language model optimized for e...
MiniMax: MiniMax M2-her
MiniMax M2-her is a dialogue-first large language model built for immersive r...
MiniMax: MiniMax M2.1
MiniMax-M2.1 is a lightweight, state-of-the-art large language model optimize...
MiniMax: MiniMax M2.5
MiniMax-M2.5 is a modified-MIT open-weights model trained with reinforcement ...
MiniMax: MiniMax M2.7
MiniMax-M2.7 is a next-generation large language model designed for autonomou...
Minimax M1 80K
Cheapest hardware per quantisation.
Each row is one quantisation tier (the same weights compressed differently). Lower precision → lower VRAM → cheaper hardware, at the cost of small accuracy loss. $/hr refreshed hourly from each provider's API.
| Quantisation | Cheapest GPU config | Total VRAM | Live $/hr | tokens/sec | |
|---|---|---|---|---|---|
|
FP16
FP16 — half precision (default)
|
1536 GB | $4.00/hr | — | Compare → | |
|
FP8
FP8 — 8-bit float (Hopper / Blackwell)
|
768 GB | $2.00/hr | — | Compare → | |
|
INT4
INT4 — 4-bit integer (~4× VRAM saving)
|
384 GB | $1.00/hr | — | Compare → |
What it costs per month across providers.
Estimate your monthly bill for MiniMax: MiniMax M3 across every host that publishes per-token pricing. Slide your token volumes; the chart + table re-rank cheapest-first.
Cheapest provider on the left.
Total monthly cost — input + output tokens combined.
Bill breakdown.
Rent the GPU instead of paying per token.
For an open-weights model like MiniMax: MiniMax M3, you can rent a GPU and serve inference yourself. The math: cheapest GPU rental × 730 hours/month + your electricity rate × power draw.
Assumes the GPU runs 24/7 at ~85% utilisation. If your traffic is bursty, you'll pay less for the API and probably more for the GPU (idle hours still cost rental). The breakeven analysis lives on the Self-host vs API breakeven tool.
About MiniMax: MiniMax M3.
MiniMax-M3 is a native multimodal model from MiniMax with roughly 428 billion parameters, about 23 billion of them active, and a 1-million-token context. It was trained on text, image and video together, and uses a sparse attention method called MiniMax Sparse Attention to reduce the work very long contexts need. A thinking setting can be enabled, disabled or left adaptive. Weights are published under the MiniMax community licence, and it targets coding and agent work.