Solar Pro Preview Instruct — Hardware Requirements & GPU Compatibility
Chat
Solar Pro Preview Instruct is a 22.1B-parameter open language model from Upstage in the Solar family. It supports a context window of up to 4,096 tokens. At BF16 it needs about 45.25 GB of VRAM — see which GPUs and Macs can run it below.
38.1K downloads
461 likesSpecifications
- Publisher
- Upstage
- Family
- Solar
- Parameters
- 22.1B
- Architecture
- SolarForCausalLM
- Context Length
- 4,096 tokens
- Vocabulary Size
- 32,128
- Release Date
- 2024-09-09
- License
- MIT
Get Started
HuggingFace
How Much VRAM Does Solar Pro Preview Instruct Need?
Select a quantization to see compatible GPUs below.
| Quantization | Bits | VRAM | + Context | File Size | Quality |
|---|---|---|---|---|---|
| BF16est. | 16.00 | 45.3 GB | 45.9 GB | 44.28 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 Solar Pro Preview Instruct?
BF16 · 45.3 GBSolar Pro Preview Instruct (BF16) requires 45.3 GB of VRAM to load the model weights. For comfortable inference with headroom for KV cache and system overhead, 59+ GB is recommended. Using the full 4K context window can add up to 0.7 GB, bringing total usage to 45.9 GB. No single GPU has enough memory — multi-GPU or cluster setups are needed.
Which Devices Can Run Solar Pro Preview Instruct?
BF16 · 45.3 GB26 devices with unified memory can run Solar Pro Preview Instruct, including NVIDIA DGX H100, NVIDIA DGX A100 640GB, Mac Studio M4 Max (64 GB).
Runs great
— Plenty of headroom
Frequently Asked Questions
- How much VRAM does Solar Pro Preview Instruct need?
Solar Pro Preview Instruct requires 45.3 GB of VRAM at BF16. Full 4K context adds up to 0.7 GB (45.9 GB total).
VRAM = Weights + KV Cache + Overhead
Weights = 22.1B × 16 bits ÷ 8 = 44.3 GB
KV Cache + Overhead ≈ 1 GB (at 2K context + ~0.3 GB framework)
KV Cache + Overhead ≈ 1.6 GB (at full 4K context)
VRAM usage by quantization
BF1645.3 GBBF16 + full context45.9 GB- Can NVIDIA GeForce RTX 5090 run Solar Pro Preview Instruct?
No — Solar Pro Preview Instruct requires at least 45.3 GB at BF16, which exceeds the NVIDIA GeForce RTX 5090's 32 GB of VRAM.
- Can I run Solar Pro Preview Instruct on a Mac?
Solar Pro Preview Instruct requires at least 45.3 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 Solar Pro Preview Instruct locally?
Yes — Solar Pro Preview Instruct can run locally on consumer hardware. At BF16 quantization it needs 45.3 GB of VRAM. Popular tools include Ollama, LM Studio, and llama.cpp.
- How fast is Solar Pro Preview Instruct?
At BF16, Solar Pro Preview Instruct can reach ~97 tok/s on AMD Instinct MI350X. 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 ÷ 45.3 × 0.65 = ~115 tok/s
Estimated speed at BF16 (45.3 GB)
~115 tok/s~115 tok/s~97 tok/sReal-world results typically within ±20%. Speed depends on batch size, quantization kernel, and software stack.
- What's the download size of Solar Pro Preview Instruct?
At BF16, the download is about 44.28 GB.
- Which GPUs can run Solar Pro Preview Instruct?
No single consumer GPU has enough VRAM to run Solar Pro Preview Instruct at BF16 (45.3 GB). Multi-GPU or professional hardware is required.
- Which devices can run Solar Pro Preview Instruct?
27 devices with unified memory can run Solar Pro Preview Instruct at BF16 (45.3 GB), including ASUS Ascent GX10, Asus ROG Flow Z13 (2025, Ryzen AI Max+ 395, 128 GB), Beelink GTR9 Pro (Ryzen AI Max+ 395, 128 GB), Framework Desktop (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.