OR

Ornith 1.5 REAP 20B A3B

OrnithGGUF by sleepyeldraziUpdated 2026-08-24
Download on Hugging Face
Parameters19B
Context256K
Layers40
Architectureqwen35moe
Smallest quant10.5 GB
Quick answer

Ornith 1.5 REAP 20B A3B needs about 15.5 GB of memory at Q5_K_M (a 14.3 GB file) with an 8K context window, so it fits on a 16 GB GPU. This repository has 5 quants, from 10.5 GB (IQ4_NL) to 38.1 GB (F16). Longer context needs more: at 32K tokens the KV cache alone adds about 2.7 GB.

Ornith 1.5 REAP 20B A3B – a lightweight, expert‑pruned GGUF model

Ornith 1.5 REAP 20B A3B is a 20.3B total‑parameter slice of the 35.9B Ornith‑1.5‑35B‑A3B. It activates ~3B parameters per token and is optimized for agentic coding and ML‑engineering tasks. The model is released under the MIT license and is available as a GGUF file.

Good at

  • Agentic coding
  • Python/C++ tooling
  • Tool‑calling

Limits

  • Requires ~16 GB VRAM for 8 k context
  • Expert pruning reduces general‑domain performance
  • No vision support

Best for

  • Local coding assistants
  • ML‑engineering debugging
  • Tool‑integrated workflows

Who it’s for. Developers and researchers who need a local, low‑memory model for coding and ML tasks. The model is suitable for users with GPUs that have at least 16 GB VRAM or Macs with 24 GB unified memory.

Quants & memory

QuantFileEst. memoryvs VRAMResultDownload
IQ4_NL10.5 GB 11.7 GB
Too big Get file →
Q5_K_M13.6 GB 14.8 GB
Too big Get file →
Q5_K_M14.3 GB 15.5 GB
Too big Get file →
F1638.1 GB 39.3 GB
Too big Get file →
F1638.1 GB 39.3 GB
Too big Get file →

For 8 GB VRAM, the IQ4_NL quant (10.48 GB file, 11.7 GB memory at 8 k context) is the lightest option. If you can afford 14 GB VRAM, use Q5_K_M (13.62 GB file, 14.8 GB memory). The F16 quant requires 38.13 GB file and 39.3 GB memory, which exceeds typical consumer GPUs.

What GPU do you need for Ornith 1.5 REAP 20B A3B?

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: the best quant that fits is IQ4_NL (10.5 GB file, about 11.7 GB in use).
  • 16 GB VRAM: the best quant that fits is Q5_K_M (14.3 GB file, about 15.5 GB in use).
  • 24 GB VRAM: the best quant that fits is Q5_K_M (14.3 GB file, about 15.5 GB in use).
  • 32 GB VRAM: the best quant that fits is Q5_K_M (14.3 GB file, about 15.5 GB in use).
  • 48 GB VRAM: the best quant that fits is F16 (38.1 GB file, about 39.3 GB in use).

Which Ornith 1.5 REAP 20B A3B quant should you choose?

Q5_K_M (14.3 GB) is the middle option in this repository. Closer to the original model than 4-bit, for a moderate increase in size. A good pick when you have memory to spare.

If memory is tight, the smallest file is IQ4_NL at 10.5 GB, with the quality trade-offs that come with fewer bits. The usual sweet spot between file size and quality, and the quant most people start with.

New to quant names? Read GGUF quantization explained.

Specifications

Architecture from the GGUF header

Architectureqwen35moe
Parameters19B (approximate, mixture of experts)
Active parameters3B
Layers40
Trained context262,144 tokens
KV cache (fp16)80 KB per token

Source

Quantized bysleepyeldrazi
FamilyOrnith
Licensemit
Downloads (HF)3,577
Last updated2026-08-24

Frequently asked questions about Ornith 1.5 REAP 20B A3B

How much VRAM does Ornith 1.5 REAP 20B A3B need?

About 15.5 GB at Q5_K_M with an 8K context window. That is the 14.3 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 Ornith 1.5 REAP 20B A3B on an 8 GB GPU?

Not at an 8K context. The smallest file needs about 11.7 GB, so you would need a larger GPU or CPU offloading.

What is the file size of Ornith 1.5 REAP 20B A3B GGUF?

It depends on the quant: IQ4_NL is 10.5 GB, Q5_K_M is 13.6 GB, Q5_K_M is 14.3 GB, F16 is 38.1 GB, F16 is 38.1 GB.

What context length does Ornith 1.5 REAP 20B A3B support?

The GGUF header lists a trained context length of 262,144 tokens (256K). The KV cache grows with the context you use: about 0.7 GB at 8K and 2.7 GB at 32K.

Where can I download Ornith 1.5 REAP 20B A3B GGUF?

The files are on Hugging Face in the sleepyeldrazi/Ornith-1.5-REAP-20B-A3B-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 Ornith 1.5 REAP 20B A3B 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.

Does the model support vision or image inputs?

No, the model has no vision capability and is focused on text and code.

What license applies to this model?

The model is released under the MIT license, allowing free use and modification.

Other GGUFs of Ornith-1.5-35B-A3B

Similar models

Browse all →

More from sleepyeldrazi

All by sleepyeldrazi →

More Ornith models

All Ornith →

Other medium models

All medium models →