SO

Models Upstage Solar Pro Preview Instruct

SolarGGUF by MaziyarPanahiUpdated 2024-09-13
Download on Hugging Face
Parameters22B
Context4K
Layers64
Architecturellama
Smallest quant4.8 GB
Quick answer

Models Upstage Solar Pro Preview Instruct needs about 16.5 GB of memory at Q4_K_M (a 13.3 GB file) with an 8K context window, so it fits on a 24 GB GPU. This repository has 16 quants, from 4.8 GB (IQ1_S) to 44.3 GB.

22‑billion‑parameter instruction‑tuned LLM for single‑GPU use

Models Upstage Solar Pro Preview Instruct is a 22 billion‑parameter llama‑based model tuned for instruction following. It supports a 4 k context window and is available in multiple GGUF quantizations. The model is released under the MIT license and is designed to run on a single GPU with 80 GB VRAM.

Good at

  • Instruction following
  • Conversational tasks
  • ChatML template usage

Limits

  • Maximum context length 4 k
  • No system prompt support
  • Limited language coverage

Best for

  • Chatbots on single‑GPU setups
  • Instruction‑driven applications
  • Rapid prototyping of conversational agents

Who it’s for. Developers and researchers needing a large instruction‑tuned model that fits on a single GPU. The model is suitable for local deployment and experimentation with the llama.cpp ecosystem.

Quants & memory

QuantFileEst. memoryvs VRAMResultDownload
IQ1_S4.8 GB 6.6 GB
Fits Get file →
IQ1_M5.2 GB 7.1 GB
Fits Get file →
IQ2_XS6.6 GB 8.5 GB
Too big Get file →
Q2_K8.2 GB 10.1 GB
Too big Get file →
IQ3_XS9.1 GB 11.0 GB
Too big Get file →
Q3_K_S9.6 GB 11.4 GB
Too big Get file →
Q3_K_M10.7 GB 12.5 GB
Too big Get file →
Q3_K_L11.6 GB 13.5 GB
Too big Get file →
IQ4_XS11.9 GB 13.7 GB
Too big Get file →
Q4_K_S12.6 GB 14.4 GB
Too big Get file →
Q4_K_M13.3 GB 15.2 GB
Too big Get file →
Q5_K_S15.2 GB 17.1 GB
Too big Get file →
Q5_K_M15.7 GB 17.5 GB
Too big Get file →
Q6_K18.2 GB 20.0 GB
Too big Get file →
Q8_023.5 GB 25.4 GB
Too big Get file →
—44.3 GB 46.1 GB
Too big Get file →

For 8 GB VRAM use IQ1_S or IQ1_M. With 16 GB VRAM, Q4_K_M or Q5_K_S offer a good balance. For 32 GB VRAM, Q6_K or Q8_0 provide higher quality while staying within memory limits. Estimates are approximate.

What GPU do you need for Models Upstage Solar Pro Preview Instruct?

These figures add the file size, an fp16 KV cache at an 8K context window (worked out from this model’s own header) and about 0.5 GB for the runtime. They are estimates, so leave a little headroom, and expect to use less context or a smaller quant if you are right at the limit.

  • 4 GB VRAM: no quant fits at an 8K context.
  • 6 GB VRAM: no quant fits at an 8K context.
  • 8 GB VRAM: the best quant that fits is IQ1_S (4.8 GB file, about 8.0 GB in use).
  • 12 GB VRAM: the best quant that fits is Q2_K (8.2 GB file, about 11.4 GB in use).
  • 16 GB VRAM: the best quant that fits is Q4_K_S (12.6 GB file, about 15.8 GB in use).
  • 24 GB VRAM: the best quant that fits is Q6_K (18.2 GB file, about 21.3 GB in use).
  • 32 GB VRAM: the best quant that fits is Q8_0 (23.5 GB file, about 26.7 GB in use).
  • 48 GB VRAM: the best quant that fits is the main file (44.3 GB file, about 47.5 GB in use).

Which Models Upstage Solar Pro Preview Instruct quant should you choose?

Q4_K_M is the usual starting point for Models Upstage Solar Pro Preview Instruct: it balances file size and quality well. Here it is a 13.3 GB file.

The Q8_0 file is 23.5 GB, about 1.8 times the size of Q4_K_M. It keeps the weights closer to the original, which only matters if you have the memory to spare.

If memory is tight, the smallest file is IQ1_S at 4.8 GB, with the quality trade-offs that come with fewer bits. Use it only when memory is very tight. Expect a clear drop in quality, which is usually worse on small models and easier to live with on large ones.

New to quant names? Read GGUF quantization explained.

Specifications

Architecture from the GGUF header

Architecturellama
Parameters22B
Layers64
Trained context4,096 tokens
KV cache (fp16)320 KB per token

Source

Quantized byMaziyarPanahi
FamilySolar
Licensemit
Downloads (HF)43,852
Last updated2024-09-13

Frequently asked questions about Models Upstage Solar Pro Preview Instruct

How much VRAM does Models Upstage Solar Pro Preview Instruct need?

About 16.5 GB at Q4_K_M with an 8K context window. That is the 13.3 GB file, plus the KV cache and roughly 0.5 GB of runtime overhead. Smaller quants need less and longer context needs more.

Can I run Models Upstage Solar Pro Preview Instruct on an 8 GB GPU?

Yes, with IQ1_S: the 4.8 GB file needs about 8.0 GB at 8K context. Larger quants in this repository do not fit in 8 GB.

What is the file size of Models Upstage Solar Pro Preview Instruct GGUF?

It depends on the quant: IQ1_S is 4.8 GB, IQ1_M is 5.2 GB, IQ2_XS is 6.6 GB, Q2_K is 8.2 GB, IQ3_XS is 9.1 GB, Q3_K_S is 9.6 GB and 10 more.

What context length does Models Upstage Solar Pro Preview Instruct support?

The GGUF header lists a trained context length of 4,096 tokens (4K). The KV cache grows with the context you use: about 2.7 GB at 8K and 1.3 GB at 4K.

Where can I download Models Upstage Solar Pro Preview Instruct GGUF?

The files are on Hugging Face in the MaziyarPanahi/solar-pro-preview-instruct-GGUF repository. The download buttons on this page open the file on Hugging Face. ggufmodels.com does not host any model files.

What license does Models Upstage Solar Pro Preview Instruct use?

Hugging Face lists the license as mit. Licenses can differ between the quantized files and the original model, so check the repository and the base model before commercial use.

Can I run Solar Pro Preview on 8 GB VRAM?

Yes, using the IQ1_S or IQ1_M quantizations, which require about 8 GB of VRAM at 8 k context.

Does the model support system prompts?

No, system prompts are not supported in the preview version.

What is the maximum context length?

The model supports a maximum context length of 4,096 tokens.

Similar models

Browse all →

More from MaziyarPanahi

All by MaziyarPanahi →

More Solar models

All Solar →

Other medium models

All medium models →