DFM Mimir — Hardware Requirements & GPU Compatibility
Chat
DFM Mimir is a 1.8B-parameter open language model from danish-foundation-models. It supports a context window of up to 4,096 tokens. At BF16 it needs about 4.07 GB of VRAM — see which GPUs and Macs can run it below.
7.4K downloads
70 likesSpecifications
- Publisher
- danish-foundation-models
- Parameters
- 1.8B
- Architecture
- HrmTextForCausalLM
- Context Length
- 4,096 tokens
- Vocabulary Size
- 262,144
- Release Date
- 2026-08-03
- License
- Apache 2.0
Get Started
HuggingFace
How Much VRAM Does DFM Mimir Need?
Select a quantization to see compatible GPUs below.
| Quantization | Bits | VRAM | + Context | File Size | Quality |
|---|---|---|---|---|---|
| BF16est. | 16.00 | 4.1 GB | 4.3 GB | 3.57 GB | Brain floating point 16 — preferred for training |
est.= calculated VRAM estimate; no published GGUF file found for that quantization yet. Other rows are verified against real community uploads.
Which GPUs Can Run DFM Mimir?
BF16 · 4.1 GBDFM Mimir (BF16) requires 4.1 GB of VRAM to load the model weights. For comfortable inference with headroom for KV cache and system overhead, 6+ GB is recommended. Using the full 4K context window can add up to 0.2 GB, bringing total usage to 4.3 GB. 50 GPUs can run it, including NVIDIA GeForce RTX 5090, NVIDIA GeForce RTX 3090 Ti.
Runs great
— Plenty of headroomWhich Devices Can Run DFM Mimir?
BF16 · 4.1 GB59 devices with unified memory can run DFM Mimir, including NVIDIA DGX H100, NVIDIA DGX A100 640GB, iPhone 17.
Runs great
— Plenty of headroom
Decent
— Enough memory, may be tightRelated Models
Frequently Asked Questions
- How much VRAM does DFM Mimir need?
DFM Mimir requires 4.1 GB of VRAM at BF16.
VRAM = Weights + KV Cache + Overhead
Weights = 1.8B × 16 bits ÷ 8 = 3.6 GB
KV Cache + Overhead ≈ 0.5 GB (at 2K context + ~0.3 GB framework)
KV Cache + Overhead ≈ 0.7 GB (at full 4K context)
VRAM usage by quantization
BF164.1 GBBF16 + full context4.3 GB- Can I run DFM Mimir on a Mac?
DFM Mimir requires at least 4.1 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 DFM Mimir locally?
Yes — DFM Mimir can run locally on consumer hardware. At BF16 quantization it needs 4.1 GB of VRAM. Popular tools include Ollama, LM Studio, and llama.cpp.
- How fast is DFM Mimir?
At BF16, DFM Mimir can reach ~1179 tok/s on AMD Instinct MI350X. On NVIDIA GeForce RTX 4090: ~161 tok/s. Speed depends mainly on GPU memory bandwidth. Real-world results typically within ±20%.
tok/s = (bandwidth GB/s ÷ model GB) × efficiency
Example: NVIDIA B200 → 8000 ÷ 4.1 × 0.65 = ~1278 tok/s
Estimated speed at BF16 (4.1 GB)
~1278 tok/s~161 tok/s~1278 tok/s~1179 tok/sReal-world results typically within ±20%. Speed depends on batch size, quantization kernel, and software stack.
- What's the download size of DFM Mimir?
At BF16, the download is about 3.57 GB.
- Which GPUs can run DFM Mimir?
50 consumer GPUs can run DFM Mimir at BF16 (4.1 GB). Top options include AMD Radeon RX 6700 XT, AMD Radeon RX 6800, AMD Radeon RX 6800 XT. 50 GPUs have plenty of headroom for comfortable inference.
- Which devices can run DFM Mimir?
59 devices with unified memory can run DFM Mimir at BF16 (4.1 GB), including AMD Ryzen AI 9 HX 370 (Strix Point) Laptop, ASUS Ascent GX10, Apple iPhone 17 Pro, Asus ROG Flow Z13 (2025, Ryzen AI Max+ 395, 128 GB). Apple Silicon Macs use unified memory shared between CPU and GPU, making them well-suited for local LLM inference.