OL

OLMoE 1B 7B 0924 Instruct

OLMoGGUF by mradermacherUpdated 2024-09-27
Download on Hugging Face
Parameters6.9B
Context4K
Layers16
Architectureolmoe
Smallest quant1.5 GB
Quick answer

OLMoE 1B 7B 0924 Instruct needs about 5.8 GB of memory at Q4_K_M (a 4.2 GB file) with an 8K context window, so it fits on a 6 GB GPU. This repository has 24 quants, from 1.5 GB (IQ1_S) to 5.7 GB (Q6_K).

OLMoE 1B 7B 0924 Instruct – a 1B active Mixture‑of‑Experts LLM

OLMoE 1B 7B 0924 Instruct is a Mixture‑of‑Experts model with 1B active and 7B total parameters. It was fine‑tuned with SFT and DPO and supports a 4096‑token context window. The model is released under the Apache‑2.0 license.

Good at

  • Instruct‑style dialogue
  • Efficient 1B parameter inference
  • Open‑source Mixture‑of‑Experts architecture

Limits

  • No vision support
  • Requires 8‑k context for memory estimates
  • Large VRAM needed for higher‑quality quants

Best for

  • Chatbots on modest GPUs
  • Rapid prototyping of instruction‑tuned LLMs
  • Research on Mixture‑of‑Experts models

Who it’s for. Developers and researchers who need a 1B‑parameter model with instruction tuning and Mixture‑of‑Experts capabilities. The model is suitable for those who prefer open‑source solutions and can manage the memory requirements of higher‑quality quantizations.

Quants & memory

QuantFileEst. memoryvs VRAMResultDownload
IQ1_S1.5 GB 2.5 GB
Fits Get file →
IQ1_M1.6 GB 2.7 GB
Fits Get file →
IQ2_XXS1.9 GB 2.9 GB
Fits Get file →
IQ2_XS2.1 GB 3.1 GB
Fits Get file →
IQ2_S2.1 GB 3.2 GB
Fits Get file →
IQ2_M2.3 GB 3.4 GB
Fits Get file →
Q2_K2.6 GB 3.6 GB
Fits Get file →
IQ3_XXS2.7 GB 3.7 GB
Fits Get file →
IQ3_XS2.9 GB 3.9 GB
Fits Get file →
IQ3_S3.0 GB 4.1 GB
Fits Get file →
Q3_K_S3.0 GB 4.1 GB
Fits Get file →
IQ3_M3.1 GB 4.1 GB
Fits Get file →
Q3_K_M3.3 GB 4.4 GB
Fits Get file →
Q3_K_L3.6 GB 4.6 GB
Fits Get file →
IQ4_XS3.7 GB 4.8 GB
Fits Get file →
Q4_03.9 GB 5.0 GB
Fits Get file →
Q4_03.9 GB 5.0 GB
Fits Get file →
Q4_03.9 GB 5.0 GB
Fits Get file →
Q4_03.9 GB 5.0 GB
Fits Get file →
Q4_K_S4.0 GB 5.0 GB
Fits Get file →
Q4_K_M4.2 GB 5.3 GB
Fits Get file →
Q5_K_S4.8 GB 5.8 GB
Fits Get file →
Q5_K_M4.9 GB 6.0 GB
Fits Get file →
Q6_K5.7 GB 6.7 GB
Fits Get file →

For 8‑GB VRAM, use IQ1_S (1.49 GB, ~3.1 GB RAM) or IQ1_M (1.64 GB, ~3.2 GB RAM). With 12‑GB VRAM, IQ2_M (2.33 GB, ~3.9 GB RAM) or Q2_K (2.56 GB, ~4.1 GB RAM) are good choices. For 24‑GB VRAM, Q5_K_M (4.93 GB, ~6.5 GB RAM) or Q6_K (5.68 GB, ~7.3 GB RAM) provide higher quality.

What GPU do you need for OLMoE 1B 7B 0924 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 IQ2_M (2.3 GB file, about 3.9 GB in use).
  • 6 GB VRAM: the best quant that fits is Q4_K_M (4.2 GB file, about 5.8 GB in use).
  • 8 GB VRAM: the best quant that fits is Q6_K (5.7 GB file, about 7.3 GB in use).

Which OLMoE 1B 7B 0924 Instruct quant should you choose?

Q4_K_M is the usual starting point for OLMoE 1B 7B 0924 Instruct: it balances file size and quality well. Here it is a 4.2 GB file.

If memory is tight, the smallest file is IQ1_S at 1.5 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

Architectureolmoe
Parameters6.9B (approximate, mixture of experts)
Layers16
Trained context4,096 tokens
KV cache (fp16)128 KB per token

Source

Quantized bymradermacher
FamilyOLMo
Licenseapache-2.0
Downloads (HF)5,500
Last updated2024-09-27

Frequently asked questions about OLMoE 1B 7B 0924 Instruct

How much VRAM does OLMoE 1B 7B 0924 Instruct need?

About 5.8 GB at Q4_K_M with an 8K context window. That is the 4.2 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 OLMoE 1B 7B 0924 Instruct on an 8 GB GPU?

Yes, with Q6_K: the 5.7 GB file needs about 7.3 GB at 8K context. Larger quants in this repository do not fit in 8 GB.

What is the file size of OLMoE 1B 7B 0924 Instruct GGUF?

It depends on the quant: IQ1_S is 1.5 GB, IQ1_M is 1.6 GB, IQ2_XXS is 1.9 GB, IQ2_XS is 2.1 GB, IQ2_S is 2.1 GB, IQ2_M is 2.3 GB and 18 more.

What context length does OLMoE 1B 7B 0924 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 1.1 GB at 8K and 0.5 GB at 4K.

Where can I download OLMoE 1B 7B 0924 Instruct GGUF?

The files are on Hugging Face in the mradermacher/OLMoE-1B-7B-0924-Instruct-i1-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 OLMoE 1B 7B 0924 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 OLMoE 1B 7B 0924 Instruct on 8 GB VRAM?

Yes, using IQ1_S or IQ1_M quantizations which require about 3.1–3.2 GB of VRAM at 8‑k context.

Does this model support image input?

No, the model does not have vision capabilities.

What license applies to this model?

The model is released under the Apache‑2.0 license.

Other GGUFs of OLMoE-1B-7B-0924-Instruct

Similar models

Browse all →

More from mradermacher

All by mradermacher →

More OLMo models

All OLMo →

Other small models

All small models →