MI

Lite-Mistral-150M-v2-Instruct

MistralGGUF by mradermacherUpdated 2024-07-14
Download on Hugging Face
Parameters157M
Context2K
Layers12
Architecturellama
Smallest quant0.1 GB
Quick answer

Lite-Mistral-150M-v2-Instruct needs about 0.8 GB of memory at Q4_K_M (a 0.1 GB file) with an 8K context window, so it fits on a 4 GB GPU. This repository has 15 quants, from 0.1 GB (Q2_K) to 0.3 GB (F16).

Lite-Mistral-150M-v2-Instruct – a compact 157 M parameter instruction model

Lite-Mistral-150M-v2-Instruct is a 157 M parameter Mistral model trained on ~8 billion tokens. It is designed for efficient operation on a wide range of devices while maintaining reasonable coherence for its size. The model may struggle with long‑term context and complex queries.

Good at

  • Short‑form instruction following
  • Simple question answering
  • Lightweight deployment

Limits

  • Limited to 2048 token context
  • May produce inconsistent responses on multi‑turn chats
  • No vision support

Best for

  • Embedded chat assistants
  • Rapid prototyping
  • Low‑resource inference

Who it’s for. Developers needing a small, instruction‑capable model for edge devices or quick experimentation. The model is open‑source under Apache‑2.0 and can be run with llama.cpp or HuggingFace transformers.

Quants & memory

QuantFileEst. memoryvs VRAMResultDownload
Q2_K0.1 GB 0.6 GB
Fits Get file →
IQ3_XS0.1 GB 0.6 GB
Fits Get file →
IQ3_S0.1 GB 0.6 GB
Fits Get file →
Q3_K_S0.1 GB 0.6 GB
Fits Get file →
IQ3_M0.1 GB 0.6 GB
Fits Get file →
Q3_K_M0.1 GB 0.6 GB
Fits Get file →
Q3_K_L0.1 GB 0.6 GB
Fits Get file →
IQ4_XS0.1 GB 0.6 GB
Fits Get file →
Q4_K_S0.1 GB 0.6 GB
Fits Get file →
Q4_K_M0.1 GB 0.6 GB
Fits Get file →
Q5_K_S0.1 GB 0.6 GB
Fits Get file →
Q5_K_M0.1 GB 0.7 GB
Fits Get file →
Q6_K0.1 GB 0.7 GB
Fits Get file →
Q8_00.2 GB 0.7 GB
Fits Get file →
F160.3 GB 0.9 GB
Fits Get file →

For 8 GB VRAM, use Q4_K_M or Q5_K_S (file size 0.1 GB, memory ~0.8 GB). For 4 GB VRAM, choose Q3_K_S or IQ3_S (0.08 GB, memory ~0.7 GB). For 2 GB VRAM, use Q2_K (0.07 GB, memory ~0.7 GB).

What GPU do you need for Lite-Mistral-150M-v2-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: the best quant that fits is F16 (0.3 GB file, about 1.0 GB in use).

Which Lite-Mistral-150M-v2-Instruct quant should you choose?

Q4_K_M is the usual starting point for Lite-Mistral-150M-v2-Instruct: it balances file size and quality well. Here it is a 0.1 GB file.

The Q8_0 file is 0.2 GB, about 1.7 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 0.1 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
Parameters157M
Layers12
Trained context2,048 tokens
KV cache (fp16)18 KB per token

Source

Quantized bymradermacher
FamilyMistral
Licenseapache-2.0
Downloads (HF)1,119
Last updated2024-07-14

Frequently asked questions about Lite-Mistral-150M-v2-Instruct

How much VRAM does Lite-Mistral-150M-v2-Instruct need?

About 0.8 GB at Q4_K_M with an 8K context window. That is the 0.1 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 Lite-Mistral-150M-v2-Instruct on an 8 GB GPU?

Yes, with F16: the 0.3 GB file needs about 1.0 GB at 8K context. Larger quants in this repository do not fit in 8 GB.

What is the file size of Lite-Mistral-150M-v2-Instruct GGUF?

It depends on the quant: Q2_K is 0.1 GB, IQ3_XS is 0.1 GB, IQ3_S is 0.1 GB, Q3_K_S is 0.1 GB, IQ3_M is 0.1 GB, Q3_K_M is 0.1 GB and 9 more.

What context length does Lite-Mistral-150M-v2-Instruct support?

The GGUF header lists a trained context length of 2,048 tokens (2K). The KV cache grows with the context you use: about 0.2 GB at 8K and 0.0 GB at 2K.

Where can I download Lite-Mistral-150M-v2-Instruct GGUF?

The files are on Hugging Face in the mradermacher/Lite-Mistral-150M-v2-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 Lite-Mistral-150M-v2-Instruct use?

Hugging Face lists the license as apache-2.0. 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 Lite-Mistral-150M-v2-Instruct on 4 GB VRAM?

Yes, using Q4_K_M or Q5_K_S quantizations will fit within 4 GB VRAM.

Does the model support vision tasks?

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

What context length does the model support?

The model is trained with a context length of 2048 tokens.

Other GGUFs of Lite-Mistral-150M-v2-Instruct

Similar models

Browse all →

More from mradermacher

All by mradermacher →

More Mistral models

All Mistral →

Other tiny models

All tiny models →