GPT OSS 20B P EAGLE needs about 2.0 GB of memory at Q4_K_M (a 1.5 GB file) with an 8K context window, so it fits on a 4 GB GPU. This repository has 3 quants, from 1.5 GB (Q4_K_M) to 3.6 GB (BF16). Longer context needs more: at 32K tokens the KV cache alone adds about 0.3 GB.
GPT OSS 20B P EAGLE – Parallel‑drafting speculative decoding model
GPT OSS 20B P EAGLE is a draft model built on the GPT‑OSS‑20B target. It uses the EAGLE3 architecture to generate 10 parallel draft tokens in a single forward pass. The model is released under Apache‑2.0 and requires a patched llama.cpp runtime.
Good at
- Parallel draft generation
- Speculative decoding
- Efficient inference with draft models
Limits
- Requires P‑EAGLE runtime patch
- No vision support
- Context length 131072 tokens
Best for
- Speed‑critical inference
- Low‑latency chat
- Batch generation
Who it’s for. Developers and researchers who need faster inference on GPT‑OSS‑20B. The model is suitable for those who can integrate the required runtime patch and manage draft‑model workflows.
Quants & memory
| Quant | File | Est. memory | vs VRAM | Result | |
|---|---|---|---|---|---|
| Q4_K_M | 1.5 GB | 2.0 GB | Fits | Get file → | |
| Q6_K | 1.9 GB | 2.4 GB | Fits | Get file → | |
| BF16 | 3.6 GB | 4.2 GB | Fits | Get file → |
For 8 GB VRAM use Q4_K_M (file size 1.46 GB, ~2 GB memory at 8 k context). For 12 GB VRAM use Q6_K (1.88 GB, ~2.4 GB). For 16 GB VRAM use BF16 (3.61 GB, ~4.2 GB). Estimates are approximate.
What GPU do you need for GPT OSS 20B P EAGLE?
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 Q6_K (1.9 GB file, about 2.4 GB in use).
- 6 GB VRAM: the best quant that fits is BF16 (3.6 GB file, about 4.2 GB in use).
Which GPT OSS 20B P EAGLE quant should you choose?
Q4_K_M is the usual starting point for GPT OSS 20B P EAGLE: it balances file size and quality well. Here it is a 1.5 GB file.
New to quant names? Read GGUF quantization explained.
Specifications
Architecture from the GGUF header
Source
Frequently asked questions about GPT OSS 20B P EAGLE
How much VRAM does GPT OSS 20B P EAGLE need?
About 2.0 GB at Q4_K_M with an 8K context window. That is the 1.5 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 GPT OSS 20B P EAGLE on an 8 GB GPU?
Yes, with BF16: the 3.6 GB file needs about 4.2 GB at 8K context. Larger quants in this repository do not fit in 8 GB.
What is the file size of GPT OSS 20B P EAGLE GGUF?
It depends on the quant: Q4_K_M is 1.5 GB, Q6_K is 1.9 GB, BF16 is 3.6 GB.
What context length does GPT OSS 20B P EAGLE support?
The GGUF header lists a trained context length of 131,072 tokens (128K). The KV cache grows with the context you use: about 0.1 GB at 8K and 0.3 GB at 32K.
Where can I download GPT OSS 20B P EAGLE GGUF?
The files are on Hugging Face in the TULLUS/GPT-OSS-20B-P-EAGLE-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 GPT OSS 20B P EAGLE 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 GPT‑OSS‑20B‑P‑EAGLE on a standard llama.cpp build?
No. A P‑EAGLE‑capable llama.cpp runtime is required; stock builds will not work.
What VRAM is needed for the Q4_K_M quantized model?
The Q4_K_M file is 1.46 GB and requires about 2 GB of VRAM at 8 k context.
Is the model multilingual?
The model is primarily English; no other languages are listed.