Wayra-Perplexity-Estimator-55M needs about 0.6 GB of memory at Q4_K_M (a 0.0 GB file) with an 8K context window, so it fits on a 4 GB GPU. This repository has 10 quants, from 0.0 GB (Q4_0) to 0.2 GB (F32).
Wayra-Perplexity-Estimator-55M: 55M‑parameter GPT‑2 model for fast perplexity estimation
Wayra-Perplexity-Estimator-55M is a 55M‑parameter GPT‑2 model optimized for high‑throughput perplexity calculation. It supports a 2048‑token context window and is released under the apache‑2.0 license. The model is available in multiple GGUF quantizations.
Good at
- Fast perplexity estimation
- High‑throughput data quality assessment
- Real‑time content filtering
Limits
- No vision support
- Requires NVIDIA A100 for TensorRT engine
- Context limited to 2048 tokens
Best for
- Large‑scale dataset cleaning
- MLOps data quality gates
- Curriculum learning ranking
Who it’s for. Data scientists and ML engineers needing rapid perplexity metrics for dataset curation or real‑time moderation. The model is suitable for production pipelines that require low latency and high throughput on compatible hardware.
Quants & memory
| Quant | File | Est. memory | vs VRAM | Result | |
|---|---|---|---|---|---|
| Q4_0 | 0.0 GB | 0.5 GB | Fits | Get file → | |
| Q4_K_M | 0.0 GB | 0.5 GB | Fits | Get file → | |
| Q4_K_S | 0.0 GB | 0.5 GB | Fits | Get file → | |
| Q5_0 | 0.0 GB | 0.5 GB | Fits | Get file → | |
| Q5_K_M | 0.0 GB | 0.6 GB | Fits | Get file → | |
| Q5_K_S | 0.0 GB | 0.6 GB | Fits | Get file → | |
| Q6_K | 0.1 GB | 0.6 GB | Fits | Get file → | |
| Q8_0 | 0.1 GB | 0.6 GB | Fits | Get file → | |
| F16 | 0.1 GB | 0.6 GB | Fits | Get file → | |
| F32 | 0.2 GB | 0.7 GB | Fits | Get file → |
For 8 GB VRAM, use Q4_0 or Q4_K_M (0.03 GB or 0.04 GB). For 12 GB VRAM, Q6_K or Q8_0 (0.06 GB) are suitable. If higher precision is needed and 16 GB VRAM is available, F16 (0.11 GB) can be used. Memory estimates are approximate.
What GPU do you need for Wayra-Perplexity-Estimator-55M?
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 F32 (0.2 GB file, about 0.8 GB in use).
Which Wayra-Perplexity-Estimator-55M quant should you choose?
Q4_K_M is the usual starting point for Wayra-Perplexity-Estimator-55M: it balances file size and quality well. Here it is a 0.0 GB file.
The Q8_0 file is 0.1 GB, about 1.6 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 Q4_0 at 0.0 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
Source
Frequently asked questions about Wayra-Perplexity-Estimator-55M
How much VRAM does Wayra-Perplexity-Estimator-55M need?
About 0.6 GB at Q4_K_M with an 8K context window. That is the 0.0 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 Wayra-Perplexity-Estimator-55M on an 8 GB GPU?
Yes, with F32: the 0.2 GB file needs about 0.8 GB at 8K context. Larger quants in this repository do not fit in 8 GB.
What is the file size of Wayra-Perplexity-Estimator-55M GGUF?
It depends on the quant: Q4_0 is 0.0 GB, Q4_K_M is 0.0 GB, Q4_K_S is 0.0 GB, Q5_0 is 0.0 GB, Q5_K_M is 0.0 GB, Q5_K_S is 0.0 GB and 4 more.
What context length does Wayra-Perplexity-Estimator-55M support?
The GGUF header lists a trained context length of 2,048 tokens (2K). The KV cache grows with the context you use: about 0.0 GB at 8K and 0.0 GB at 2K.
Where can I download Wayra-Perplexity-Estimator-55M GGUF?
The files are on Hugging Face in the latam-gpt/Wayra-Perplexity-Estimator-55M repository. The download buttons on this page open the file on Hugging Face. ggufmodels.com does not host any model files.
What license does Wayra-Perplexity-Estimator-55M 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 this model on 8 GB VRAM?
Yes, the Q4_0 or Q4_K_M quantizations are under 0.1 GB and use about 0.6 GB of GPU memory at 8k context.
Does the model support vision tasks?
No, the model has no vision capabilities and is text‑only.
What hardware is required for the TensorRT engine?
The TensorRT engine is A100‑optimized and requires sm_80 GPUs with CUDA 12.8+ and TensorRT 10.13.x.