Zephyr Orpo 141B A35b v0.1 — Hardware Requirements & GPU Compatibility
Chat
Zephyr Orpo 141B A35b v0.1 is a 140.6B-parameter open language model from Hugging Face. It supports a context window of up to 65,536 tokens. At BF16 it needs about 282.01 GB of VRAM — see which GPUs and Macs can run it below.
158 downloads
269 likesSpecifications
- Publisher
- Hugging Face
- Parameters
- 140.6B
- Architecture
- MixtralForCausalLM
- Context Length
- 65,536 tokens
- Vocabulary Size
- 32,000
- Release Date
- 2024-04-18
- License
- Apache 2.0
Get Started
HuggingFace
How Much VRAM Does Zephyr Orpo 141B A35b v0.1 Need?
Select a quantization to see compatible GPUs below.
| Quantization | Bits | VRAM | + Context | File Size | Quality |
|---|---|---|---|---|---|
| BF16 | 16.00 | 282.0 GB | 296.6 GB | 281.24 GB | Brain floating point 16 — preferred for training |
Which GPUs Can Run Zephyr Orpo 141B A35b v0.1?
BF16 · 282.0 GBZephyr Orpo 141B A35b v0.1 (BF16) requires 282.0 GB of VRAM to load the model weights. For comfortable inference with headroom for KV cache and system overhead, 367+ GB is recommended. Using the full 66K context window can add up to 14.6 GB, bringing total usage to 296.6 GB. No single GPU has enough memory — multi-GPU or cluster setups are needed.
Which Devices Can Run Zephyr Orpo 141B A35b v0.1?
BF16 · 282.0 GB2 devices with unified memory can run Zephyr Orpo 141B A35b v0.1, including NVIDIA DGX H100, NVIDIA DGX A100 640GB.
Runs great
— Plenty of headroomRelated Models
Frequently Asked Questions
- How much VRAM does Zephyr Orpo 141B A35b v0.1 need?
Zephyr Orpo 141B A35b v0.1 requires 282.0 GB of VRAM at BF16. Full 66K context adds up to 14.6 GB (296.6 GB total).
VRAM = Weights + KV Cache + Overhead
Weights = 140.6B × 16 bits ÷ 8 = 281.2 GB
KV Cache + Overhead ≈ 0.8 GB (at 2K context + ~0.3 GB framework)
KV Cache + Overhead ≈ 15.4 GB (at full 66K context)
VRAM usage by quantization
BF16282.0 GBBF16 + full context296.6 GB- Can NVIDIA GeForce RTX 5090 run Zephyr Orpo 141B A35b v0.1?
No — Zephyr Orpo 141B A35b v0.1 requires at least 282.0 GB at BF16, which exceeds the NVIDIA GeForce RTX 5090's 32 GB of VRAM.
- Can I run Zephyr Orpo 141B A35b v0.1 on a Mac?
Zephyr Orpo 141B A35b v0.1 requires at least 282.0 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 Zephyr Orpo 141B A35b v0.1 locally?
Yes — Zephyr Orpo 141B A35b v0.1 can run locally on consumer hardware. At BF16 quantization it needs 282.0 GB of VRAM. Popular tools include Ollama, LM Studio, and llama.cpp.
- What's the download size of Zephyr Orpo 141B A35b v0.1?
At BF16, the download is about 281.24 GB.
- Which GPUs can run Zephyr Orpo 141B A35b v0.1?
No single consumer GPU has enough VRAM to run Zephyr Orpo 141B A35b v0.1 at BF16 (282.0 GB). Multi-GPU or professional hardware is required.
- Which devices can run Zephyr Orpo 141B A35b v0.1?
2 devices with unified memory can run Zephyr Orpo 141B A35b v0.1 at BF16 (282.0 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.