MI

Noromaid v0.4 Mixtral Instruct 8x7b Zloss

MistralGGUF by mradermacherUpdated 2024-05-06
Download on Hugging Face
Parameters46.7B
Context32K
Layers32
Architecturellama
Smallest quant17.7 GB
Quick answer

Noromaid v0.4 Mixtral Instruct 8x7b Zloss needs about 30.4 GB of memory at Q4_K_M (a 28.9 GB file) with an 8K context window, so it fits on a 32 GB GPU. This repository has 13 quants, from 17.7 GB (Q2_K) to 38.8 GB (Q6_K). Longer context needs more: at 32K tokens the KV cache alone adds about 4.3 GB.

Noromaid v0.4 Mixtral Instruct 8x7b Zloss – 46.7B parameter Mistral model

Noromaid v0.4 Mixtral Instruct 8x7b Zloss is a 46.7B parameter Mistral model trained on a 32768 token context. It uses Chatml prompting and is available in multiple GGUF quantizations. The model is licensed under cc‑by‑nc‑4.0.

Good at

  • Chatml prompt handling
  • Large context reasoning
  • Instruction following

Limits

  • Experimental, not all features guaranteed
  • No vision support
  • Requires 8k context for memory estimates

Best for

  • Role‑playing scenarios
  • Creative writing prompts
  • Instructional dialogue

Who it’s for. Developers and researchers needing a large Mistral model with Chatml support. The model is suitable for experimentation and fine‑tuning within the cc‑by‑nc‑4.0 license.

Quants & memory

QuantFileEst. memoryvs VRAMResultDownload
Q2_K17.7 GB 19.3 GB
Too big Get file →
IQ3_XS19.7 GB 21.3 GB
Too big Get file →
IQ3_S20.8 GB 22.4 GB
Too big Get file →
Q3_K_S20.8 GB 22.4 GB
Too big Get file →
IQ3_M21.8 GB 23.4 GB
Too big Get file →
Q3_K_M23.0 GB 24.5 GB
Too big Get file →
Q3_K_L24.6 GB 26.2 GB
Too big Get file →
IQ4_XS25.8 GB 27.4 GB
Too big Get file →
Q4_K_S27.2 GB 28.7 GB
Too big Get file →
Q4_K_M28.9 GB 30.4 GB
Too big Get file →
Q5_K_S32.6 GB 34.2 GB
Too big Get file →
Q5_K_M33.6 GB 35.2 GB
Too big Get file →
Q6_K38.8 GB 40.4 GB
Too big Get file →

For 8 GB VRAM, use Q2_K (file size 17.73 GB, estimated 19.3 GB memory). For 16 GB VRAM, Q4_K_S (27.16 GB, 28.7 GB memory) or Q3_K_M (22.96 GB, 24.5 GB memory). For 24 GB VRAM, Q5_K_M (33.65 GB, 35.2 GB memory) or Q6_K (38.8 GB, 40.4 GB memory). Estimates are approximate.

What GPU do you need for Noromaid v0.4 Mixtral Instruct 8x7b Zloss?

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: no quant fits at an 8K context.
  • 12 GB VRAM: no quant fits at an 8K context.
  • 16 GB VRAM: no quant fits at an 8K context.
  • 24 GB VRAM: the best quant that fits is IQ3_M (21.8 GB file, about 23.4 GB in use).
  • 32 GB VRAM: the best quant that fits is Q4_K_M (28.9 GB file, about 30.4 GB in use).
  • 48 GB VRAM: the best quant that fits is Q6_K (38.8 GB file, about 40.4 GB in use).

Which Noromaid v0.4 Mixtral Instruct 8x7b Zloss quant should you choose?

Q4_K_M is the usual starting point for Noromaid v0.4 Mixtral Instruct 8x7b Zloss: it balances file size and quality well. Here it is a 28.9 GB file.

If memory is tight, the smallest file is Q2_K at 17.7 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
Parameters46.7B (approximate, mixture of experts)
Layers32
Trained context32,768 tokens
KV cache (fp16)128 KB per token

Frequently asked questions about Noromaid v0.4 Mixtral Instruct 8x7b Zloss

How much VRAM does Noromaid v0.4 Mixtral Instruct 8x7b Zloss need?

About 30.4 GB at Q4_K_M with an 8K context window. That is the 28.9 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 Noromaid v0.4 Mixtral Instruct 8x7b Zloss on an 8 GB GPU?

Not at an 8K context. The smallest file needs about 19.3 GB, so you would need a larger GPU or CPU offloading.

What is the file size of Noromaid v0.4 Mixtral Instruct 8x7b Zloss GGUF?

It depends on the quant: Q2_K is 17.7 GB, IQ3_XS is 19.7 GB, IQ3_S is 20.8 GB, Q3_K_S is 20.8 GB, IQ3_M is 21.8 GB, Q3_K_M is 23.0 GB and 7 more.

What context length does Noromaid v0.4 Mixtral Instruct 8x7b Zloss support?

The GGUF header lists a trained context length of 32,768 tokens (32K). The KV cache grows with the context you use: about 1.1 GB at 8K and 4.3 GB at 32K.

Where can I download Noromaid v0.4 Mixtral Instruct 8x7b Zloss GGUF?

The files are on Hugging Face in the mradermacher/Noromaid-v0.4-Mixtral-Instruct-8x7b-Zloss-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 Noromaid v0.4 Mixtral Instruct 8x7b Zloss use?

Hugging Face lists the license as cc-by-nc-4.0. 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 this model on 8 GB VRAM?

Use the Q2_K quant; it requires about 19.3 GB of VRAM at 8k context, so 8 GB is insufficient.

Does this model support vision?

No, the model has no vision capabilities.

What license applies to this model?

The model is licensed under cc‑by‑nc‑4.0, allowing non‑commercial use with attribution.

Other GGUFs of Noromaid-v0.4-Mixtral-Instruct-8x7b-Zloss

Similar models

Browse all →

More from mradermacher

All by mradermacher →

More Mistral models

All Mistral →

Other large models

All large models →