Darwin 218B Delphi — Hardware Requirements & GPU Compatibility
Chat
Darwin 218B Delphi is a 218.8B-parameter open language model from FINAL-Bench in the Phi family. It supports a context window of up to 5,000,000 tokens. At BF16 it needs about 437.87 GB of VRAM — see which GPUs and Macs can run it below.
24 downloads
3 likesSpecifications
- Publisher
- FINAL-Bench
- Family
- Phi
- Parameters
- 218.8B
- Architecture
- Cohere2VisionForConditionalGeneration
- Context Length
- 5,000,000 tokens
- Vocabulary Size
- 262,144
- Release Date
- 2026-06-03
- License
- Apache 2.0
Get Started
HuggingFace
How Much VRAM Does Darwin 218B Delphi Need?
Select a quantization to see compatible GPUs below.
| Quantization | Bits | VRAM | + Context | File Size | Quality |
|---|---|---|---|---|---|
| BF16 | 16.00 | 437.9 GB | 601.6 GB | 437.50 GB | Brain floating point 16 — preferred for training |
Which GPUs Can Run Darwin 218B Delphi?
BF16 · 437.9 GBDarwin 218B Delphi (BF16) requires 437.9 GB of VRAM to load the model weights. For comfortable inference with headroom for KV cache and system overhead, 570+ GB is recommended. Using the full 5000K context window can add up to 163.8 GB, bringing total usage to 601.6 GB. No single GPU has enough memory — multi-GPU or cluster setups are needed.
Which Devices Can Run Darwin 218B Delphi?
BF16 · 437.9 GB2 devices with unified memory can run Darwin 218B Delphi, including NVIDIA DGX H100.
Decent
— Enough memory, may be tightRelated Models
Frequently Asked Questions
- How much VRAM does Darwin 218B Delphi need?
Darwin 218B Delphi requires 437.9 GB of VRAM at BF16. Full 5000K context adds up to 163.8 GB (601.6 GB total).
VRAM = Weights + KV Cache + Overhead
Weights = 218.8B × 16 bits ÷ 8 = 437.5 GB
KV Cache + Overhead ≈ 0.4 GB (at 2K context + ~0.3 GB framework)
KV Cache + Overhead ≈ 164.1 GB (at full 5000K context)
VRAM usage by quantization
BF16437.9 GBBF16 + full context601.6 GB- Can NVIDIA GeForce RTX 5090 run Darwin 218B Delphi?
No — Darwin 218B Delphi requires at least 437.9 GB at BF16, which exceeds the NVIDIA GeForce RTX 5090's 32 GB of VRAM.
- Can I run Darwin 218B Delphi on a Mac?
Darwin 218B Delphi requires at least 437.9 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 Darwin 218B Delphi locally?
Yes — Darwin 218B Delphi can run locally on consumer hardware. At BF16 quantization it needs 437.9 GB of VRAM. Popular tools include Ollama, LM Studio, and llama.cpp.
- What's the download size of Darwin 218B Delphi?
At BF16, the download is about 437.50 GB.
- Which GPUs can run Darwin 218B Delphi?
No single consumer GPU has enough VRAM to run Darwin 218B Delphi at BF16 (437.9 GB). Multi-GPU or professional hardware is required.
- Which devices can run Darwin 218B Delphi?
2 devices with unified memory can run Darwin 218B Delphi at BF16 (437.9 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.