SO

SoMix2-xb

SomixGGUF by mradermacherUpdated 2024-06-12
Download on Hugging Face
Parameters11.2B
Context32K
Layers50
Architecturellama
Smallest quant4.2 GB
Quick answer

SoMix2-xb needs about 8.9 GB of memory at Q4_K_M (a 6.7 GB file) with an 8K context window, so it fits on a 12 GB GPU. This repository has 14 quants, from 4.2 GB (Q2_K) to 11.9 GB (Q8_0). Longer context needs more: at 32K tokens the KV cache alone adds about 6.7 GB.

SoMix2-xb – 11.2B llama model in GGUF format

SoMix2-xb is an 11.2B parameter llama model quantized into GGUF files. It was created by merging TheTop-5x7B-Instruct-S3-v0.1 and notus-7b-v1 using LazyMergekit. The model supports a context length of 32768 tokens.

Good at

  • Large context handling
  • Instruction following
  • Multi‑domain text generation

Limits

  • No vision support
  • License not specified
  • Requires 8k context for memory estimates

Best for

  • Chatbots needing long context
  • Research prototyping
  • Content generation with extended memory

Who it’s for. Developers and researchers who need a large‑parameter llama model with extended context in a lightweight GGUF format. The model is suitable for local deployment on machines with sufficient VRAM and RAM.

Quants & memory

QuantFileEst. memoryvs VRAMResultDownload
Q2_K4.2 GB 6.3 GB
Fits Get file →
IQ3_XS4.6 GB 6.8 GB
Fits Get file →
Q3_K_S4.9 GB 7.0 GB
Fits Get file →
IQ3_S4.9 GB 7.1 GB
Fits Get file →
IQ3_M5.0 GB 7.2 GB
Tight Get file →
Q3_K_M5.4 GB 7.6 GB
Tight Get file →
Q3_K_L5.9 GB 8.1 GB
Too big Get file →
IQ4_XS6.1 GB 8.2 GB
Too big Get file →
Q4_K_S6.4 GB 8.5 GB
Too big Get file →
Q4_K_M6.7 GB 8.9 GB
Too big Get file →
Q5_K_S7.7 GB 9.9 GB
Too big Get file →
Q5_K_M7.9 GB 10.1 GB
Too big Get file →
Q6_K9.2 GB 11.3 GB
Too big Get file →
Q8_011.9 GB 14.0 GB
Too big Get file →

For 8‑GB VRAM, the Q4_K_S quant (~6.36 GB file, ~8.5 GB memory at 8k context) is a good balance. If 12‑GB VRAM is available, Q5_K_S (~7.7 GB file, ~9.9 GB memory) offers higher quality. For 16‑GB VRAM, Q6_K (~9.16 GB file, ~11.3 GB memory) provides the best performance.

What GPU do you need for SoMix2-xb?

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 Q3_K_M (5.4 GB file, about 7.6 GB in use).
  • 12 GB VRAM: the best quant that fits is Q6_K (9.2 GB file, about 11.3 GB in use).
  • 16 GB VRAM: the best quant that fits is Q8_0 (11.9 GB file, about 14.0 GB in use).

Which SoMix2-xb quant should you choose?

Q4_K_M is the usual starting point for SoMix2-xb: it balances file size and quality well. Here it is a 6.7 GB file.

The Q8_0 file is 11.9 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 Q2_K at 4.2 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
Parameters11.2B
Layers50
Trained context32,768 tokens
KV cache (fp16)200 KB per token

Source

Quantized bymradermacher
FamilySomix
Downloads (HF)1,184
Last updated2024-06-12

Frequently asked questions about SoMix2-xb

How much VRAM does SoMix2-xb need?

About 8.9 GB at Q4_K_M with an 8K context window. That is the 6.7 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 SoMix2-xb on an 8 GB GPU?

Yes, with Q3_K_M: the 5.4 GB file needs about 7.6 GB at 8K context. Larger quants in this repository do not fit in 8 GB.

What is the file size of SoMix2-xb GGUF?

It depends on the quant: Q2_K is 4.2 GB, IQ3_XS is 4.6 GB, Q3_K_S is 4.9 GB, IQ3_S is 4.9 GB, IQ3_M is 5.0 GB, Q3_K_M is 5.4 GB and 8 more.

What context length does SoMix2-xb support?

The GGUF header lists a trained context length of 32,768 tokens (32K). The KV cache grows with the context you use: about 1.7 GB at 8K and 6.7 GB at 32K.

Where can I download SoMix2-xb GGUF?

The files are on Hugging Face in the mradermacher/SoMix2-xb-GGUF repository. The download buttons on this page open the file on Hugging Face. ggufmodels.com does not host any model files.

Can I run SoMix2-xb on 8 GB VRAM?

The Q4_K_S quant uses about 8.5 GB memory at 8k context, so 8 GB VRAM is insufficient. Use a lower‑bit quant like Q2_K or Q3_K_S if you need to fit in 8 GB.

What is the maximum context length supported?

The model was trained with 32768 tokens and supports that context length in inference.

Does SoMix2-xb support vision tasks?

No, the model has no vision capabilities and is text‑only.

Similar models

Browse all →

More from mradermacher

All by mradermacher →

More Somix models

All Somix →

Other medium models

All medium models →