MiniMax-M3
MiniMax-M3 is a native multimodal model with 1M context. It has ~428B parameters and ~23B activated parameters.
Params
427.0 B
Context
—
Downloads 30d
203 K
Likes
1,594
Download history
tracking started — chart appears after 7 days of snapshots (5 recorded)203 K downloads in the last 30 days
Can you run it?
Estimated VRAM at 8K context unless noted. Pick your hardware to see the verdict per quantization.
| File | Quant | Size | Est. VRAM | Verdict on RTX 4090 · 24 GB |
|---|---|---|---|---|
| model.safetensors | bf16 | 854.2 GB | 1,004.2 GB | ❌ Won’t fit |
Estimate: file size × 1.1 + KV cache at 8K + 0.5 GB overhead. Not a benchmark — how we calculate this.
Use via API
via OpenRouter · USD per 1M tokens · synced 2026-10-24$0.30
input / 1M
$1.20
output / 1M
$0.06
cached input / 1M
$1.50
1M in + 1M out
- Context window
- 524,288 tokens
- Max output
- 512,000 tokens
- Providers
- 12
- Cheapest provider
- DeepInfra · $0.22 / $0.88
- Best throughput
- —
- AA Intelligence / Coding / Agentic
- 29.2 / 58.6 / 29.5
✓ Tool calling
✓ Reasoning
✓ Structured outputs
✓ Image input
✗ File input
✗ Audio input
✓ Prompt caching
Specifications
- Architecture
- MiniMaxM3SparseForConditionalGeneration
- Parameters
- 427.0 B
- Tensor type
- BF16
- Licence
- other
- First seen on the Hub
- 2026-06-02
- Training datasets
- undisclosed
- Added to our catalog
- 2026-10-20
Compare with any image-text-to-text model