EuroLLM 9B — Hardware Requirements & GPU Compatibility
Chat
2.2K downloads
161 likesSpecifications
- Publisher
- utter-project
- Parameters
- 9.2B
- Release Date
- 2024-12-09
- License
- Apache 2.0
Get Started
HuggingFace
How Much VRAM Does EuroLLM 9B Need?
Select a quantization to see compatible GPUs below.
| Quantization | Bits | VRAM | + Context | File Size | Quality |
|---|---|---|---|---|---|
| BF16 | 16.00 | 20.1 GB | — | 18.30 GB | Brain floating point 16 — preferred for training |
Which GPUs Can Run EuroLLM 9B?
BF16 · 20.1 GBEuroLLM 9B (BF16) requires 20.1 GB of VRAM to load the model weights. For comfortable inference with headroom for KV cache and system overhead, 27+ GB is recommended. 5 GPUs can run it, including NVIDIA GeForce RTX 5090, NVIDIA GeForce RTX 3090 Ti.
Runs great
— Plenty of headroom
Decent
— Enough VRAM, may be tightWhich Devices Can Run EuroLLM 9B?
BF16 · 20.1 GB21 devices with unified memory can run EuroLLM 9B, including NVIDIA DGX H100, NVIDIA DGX A100 640GB, Mac Mini M4 Pro (24 GB).
Runs great
— Plenty of headroomRelated Models
Frequently Asked Questions
- How much VRAM does EuroLLM 9B need?
EuroLLM 9B requires 20.1 GB of VRAM at BF16.
VRAM = Weights + KV Cache + Overhead
Weights = 9.2B × 16 bits ÷ 8 = 18.3 GB
KV Cache + Overhead ≈ 1.8 GB (at 2K context + ~0.3 GB framework)
VRAM usage by quantization
BF1620.1 GB- Can I run EuroLLM 9B on a Mac?
EuroLLM 9B requires at least 20.1 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 EuroLLM 9B locally?
Yes — EuroLLM 9B can run locally on consumer hardware. At BF16 quantization it needs 20.1 GB of VRAM. Popular tools include Ollama, LM Studio, and llama.cpp.
- How fast is EuroLLM 9B?
At BF16, EuroLLM 9B can reach ~145 tok/s on AMD Instinct MI300X. On NVIDIA GeForce RTX 4090: ~33 tok/s. Speed depends mainly on GPU memory bandwidth. Real-world results typically within ±20%.
tok/s = (bandwidth GB/s ÷ model GB) × efficiency
Example: AMD Instinct MI300X → 5300 ÷ 20.1 × 0.55 = ~145 tok/s
Estimated speed at BF16 (20.1 GB)
AMD Instinct MI300X~145 tok/sNVIDIA GeForce RTX 4090~33 tok/sNVIDIA H100 SXM~108 tok/sAMD Instinct MI250X~90 tok/sReal-world results typically within ±20%. Speed depends on batch size, quantization kernel, and software stack.
- What's the download size of EuroLLM 9B?
At BF16, the download is about 18.30 GB.
- Which GPUs can run EuroLLM 9B?
5 consumer GPUs can run EuroLLM 9B at BF16 (20.1 GB). Top options include NVIDIA GeForce RTX 5090, AMD Radeon RX 7900 XTX, NVIDIA GeForce RTX 3090. 1 GPU have plenty of headroom for comfortable inference.
- Which devices can run EuroLLM 9B?
21 devices with unified memory can run EuroLLM 9B at BF16 (20.1 GB), including Mac Mini M4 (32 GB), Mac Mini M4 Pro (24 GB), Mac Mini M4 Pro (48 GB), Mac Pro M2 Ultra (192 GB). Apple Silicon Macs use unified memory shared between CPU and GPU, making them well-suited for local LLM inference.