MiroThinker 1.7 — Hardware Requirements & GPU Compatibility
ChatFunctions
158 downloads
59 likesSpecifications
- Publisher
- miromind-ai
- Parameters
- 235.1B
- Architecture
- Qwen3MoeForCausalLM
- Context Length
- 262,144 tokens
- Vocabulary Size
- 151,936
- Release Date
- 2026-03-12
- License
- Apache 2.0
Get Started
HuggingFace
How Much VRAM Does MiroThinker 1.7 Need?
Select a quantization to see compatible GPUs below.
| Quantization | Bits | VRAM | + Context | File Size | Quality |
|---|---|---|---|---|---|
| BF16 | 16.00 | 470.7 GB | 495.7 GB | 470.19 GB | Brain floating point 16 — preferred for training |
Which GPUs Can Run MiroThinker 1.7?
BF16 · 470.7 GBMiroThinker 1.7 (BF16) requires 470.7 GB of VRAM to load the model weights. For comfortable inference with headroom for KV cache and system overhead, 612+ GB is recommended. Using the full 262K context window can add up to 25.0 GB, bringing total usage to 495.7 GB. No single GPU has enough memory — multi-GPU or cluster setups are needed.
Which Devices Can Run MiroThinker 1.7?
BF16 · 470.7 GB2 devices with unified memory can run MiroThinker 1.7, including NVIDIA DGX H100.
Decent
— Enough memory, may be tightRelated Models
Frequently Asked Questions
- How much VRAM does MiroThinker 1.7 need?
MiroThinker 1.7 requires 470.7 GB of VRAM at BF16. Full 262K context adds up to 25.0 GB (495.7 GB total).
VRAM = Weights + KV Cache + Overhead
Weights = 235.1B × 16 bits ÷ 8 = 470.2 GB
KV Cache + Overhead ≈ 0.5 GB (at 2K context + ~0.3 GB framework)
KV Cache + Overhead ≈ 25.5 GB (at full 262K context)
VRAM usage by quantization
BF16470.7 GBBF16 + full context495.7 GB- Can NVIDIA GeForce RTX 5090 run MiroThinker 1.7?
No — MiroThinker 1.7 requires at least 470.7 GB at BF16, which exceeds the NVIDIA GeForce RTX 5090's 32 GB of VRAM.
- Can I run MiroThinker 1.7 on a Mac?
MiroThinker 1.7 requires at least 470.7 GB at BF16, which exceeds the unified memory of most consumer Macs. You would need a Mac Studio or Mac Pro with a high-memory configuration.
- Can I run MiroThinker 1.7 locally?
Yes — MiroThinker 1.7 can run locally on consumer hardware. At BF16 quantization it needs 470.7 GB of VRAM. Popular tools include Ollama, LM Studio, and llama.cpp.
- What's the download size of MiroThinker 1.7?
At BF16, the download is about 470.19 GB.
- Which GPUs can run MiroThinker 1.7?
No single consumer GPU has enough VRAM to run MiroThinker 1.7 at BF16 (470.7 GB). Multi-GPU or professional hardware is required.
- Which devices can run MiroThinker 1.7?
2 devices with unified memory can run MiroThinker 1.7 at BF16 (470.7 GB), including NVIDIA DGX A100 640GB, NVIDIA DGX H100. Apple Silicon Macs use unified memory shared between CPU and GPU, making them well-suited for local LLM inference.