EagleVLA needs about 6.9 GB of memory at its main file (a 6.1 GB file) with an 8K context window, so it fits on a 8 GB GPU. This repository has 3 quants, from 5.7 GB to 6.1 GB (BF16). Longer context needs more: at 32K tokens the KV cache alone adds about 1.2 GB.
EagleVLA – a 2.8B vision‑language‑action foundation model
EagleVLA is a 2.8B parameter VLA model based on π₀. It accepts images, proprioceptive data, and optional language instructions to output continuous actions. The model is designed for fine‑tuning on specific robotic tasks.
Good at
- Vision‑language reasoning
- Continuous action generation
- Real‑time inference on CUDA platforms
Limits
- No explicit license information
- No vision support beyond the base model
- Training context limited to 131072 tokens
Best for
- Robot control pipelines
- Fine‑tuning for custom action tasks
- Embedded inference on Jetson devices
Who it’s for. Engineers and researchers building vision‑language‑action systems who need a lightweight base model for fine‑tuning. The model is suitable for developers targeting CUDA‑capable hardware such as NVIDIA Jetson.
Quants & memory
| Quant | File | Est. memory | vs VRAM | Result | |
|---|---|---|---|---|---|
| — | 5.7 GB | 6.5 GB | Fits | Get file → | |
| — | 6.1 GB | 6.9 GB | Fits | Get file → | |
| BF16 | 6.1 GB | 6.9 GB | Fits | Get file → |
For 8 GB VRAM, the 6.12 GB file with an estimated 6.9 GB memory at 8 k context is suitable. If memory is tighter, the 5.66 GB file uses about 6.5 GB at 8 k context. The BF16 quant is 6.14 GB and also requires roughly 6.9 GB memory.
What GPU do you need for EagleVLA?
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: the best quant that fits is BF16 (6.1 GB file, about 6.9 GB in use).
Which EagleVLA quant should you choose?
If memory is tight, the smallest file is the first one in the table at 5.7 GB, with the quality trade-offs that come with fewer bits.
New to quant names? Read GGUF quantization explained.
Specifications
Architecture from the GGUF header
Source
Frequently asked questions about EagleVLA
How much VRAM does EagleVLA need?
About 6.9 GB at its main file with an 8K context window. That is the 6.1 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 EagleVLA on an 8 GB GPU?
Yes, with BF16: the 6.1 GB file needs about 6.9 GB at 8K context. Larger quants in this repository do not fit in 8 GB.
What is the file size of EagleVLA GGUF?
It depends on the quant: file is 5.7 GB, file is 6.1 GB, BF16 is 6.1 GB.
What context length does EagleVLA support?
The GGUF header lists a trained context length of 131,072 tokens (128K). The KV cache grows with the context you use: about 0.3 GB at 8K and 1.2 GB at 32K.
Where can I download EagleVLA GGUF?
The files are on Hugging Face in the diantoudefengshan/EagleVLA-GGUF repository. The download buttons on this page open the file on Hugging Face. ggufmodels.com does not host any model files.
Can I run EagleVLA on 8 GB VRAM?
Yes. The 6.12 GB quant file uses about 6.9 GB memory at 8 k context, fitting within 8 GB VRAM.
What is the maximum context length for this model?
The trained context tokens are 131072, but inference typically uses 8 k context.
Does EagleVLA support vision inputs?
Yes, the model is designed for vision‑language‑action tasks and accepts images as input.