Xing4.0 29B A4B APEX needs about 21.9 GB of memory at its main file (a 20.6 GB file) with an 8K context window, so it fits on a 24 GB GPU. This repository has 3 quants, from 15.0 GB to 22.8 GB. Longer context needs more: at 32K tokens the KV cache alone adds about 2.9 GB.
Xing4.0 29B A4B APEX – 31.2B parameter LLM with 4B active MoE
Xing4.0 29B A4B APEX is a 31.2B parameter model with only 4B activated per token. It supports a 262144 token context length and is trained on the Ascend NPU platform. The model is released under Apache‑2.0 and is compatible with llama.cpp, vLLM, SGLang, and KTransformers.
Good at
- Reasoning
- Tool calling
- Long‑context tasks
Limits
- No vision support
- Requires large VRAM
- Licensed under Apache‑2.0
Best for
- Complex engineering queries
- Agent‑oriented workflows
- Domain‑specific fine‑tuning
Who it’s for. Developers and researchers needing a large‑parameter model with efficient MoE activation for long‑context reasoning. The model is suitable for teams that can deploy on Ascend or GPU clusters and are comfortable with open‑source inference frameworks.
Quants & memory
| Quant | File | Est. memory | vs VRAM | Result | |
|---|---|---|---|---|---|
| — | 15.0 GB | 16.3 GB | Too big | Get file → | |
| — | 20.6 GB | 21.9 GB | Too big | Get file → | |
| — | 22.8 GB | 24.0 GB | Too big | Get file → |
For 8 GB VRAM use the 15.03 GB file (≈16.3 GB memory at 8 k context). For 12 GB VRAM use the 20.63 GB file (≈21.9 GB memory). For 16 GB VRAM use the 22.75 GB file (≈24 GB memory). These estimates are approximate.
What GPU do you need for Xing4.0 29B A4B APEX?
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: no quant fits at an 8K context.
- 16 GB VRAM: no quant fits at an 8K context.
- 24 GB VRAM: the best quant that fits is the main file (22.8 GB file, about 24.0 GB in use).
Which Xing4.0 29B A4B APEX quant should you choose?
If memory is tight, the smallest file is the first one in the table at 15.0 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 Xing4.0 29B A4B APEX
How much VRAM does Xing4.0 29B A4B APEX need?
About 21.9 GB at its main file with an 8K context window. That is the 20.6 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 Xing4.0 29B A4B APEX on an 8 GB GPU?
Not at an 8K context. The smallest file needs about 16.3 GB, so you would need a larger GPU or CPU offloading.
What is the file size of Xing4.0 29B A4B APEX GGUF?
It depends on the quant: file is 15.0 GB, file is 20.6 GB, file is 22.8 GB.
What context length does Xing4.0 29B A4B APEX 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 2.9 GB at 32K.
Where can I download Xing4.0 29B A4B APEX GGUF?
The files are on Hugging Face in the vincespeed/Xing4.0-29B-A4B-APEX-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 Xing4.0 29B A4B APEX 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 Xing4.0 29B A4B on 8 GB VRAM?
The smallest quantized file is 15.03 GB, which requires about 16.3 GB of VRAM at 8 k context, so 8 GB VRAM is insufficient.
Does this model support vision tasks?
No, the model has no vision capability.
What license does the model use?
The model is released under the Apache‑2.0 license.