qwen2.5 coder 32b bitnet 1.58b needs about 13.8 GB of memory at its main file (a 11.2 GB file) with an 8K context window, so it fits on a 16 GB GPU. Longer context needs more: at 32K tokens the KV cache alone adds about 8.6 GB.
Qwen2.5-Coder 32B BitNet 1.58b – 32B parameter code LLM
This model is a 32.8B parameter Qwen2.5-Coder architecture quantized with BitNet 1.58-bit ternary quantization. It supports up to 32,768 context tokens and is trained for code generation, reasoning, and fixing. The GGUF file size is 11.17 GB with an estimated VRAM usage of 13.8 GB at 8 k context.
Good at
- Code generation
- Code reasoning
- Code fixing
Limits
- No vision support
- Quality not benchmarked
- Requires custom runtime
Best for
- Developers needing large code models
- Code agents
- Long‑context code analysis
Who it’s for. This model is intended for developers and researchers who need a large code‑focused LLM and are comfortable with experimental quantization. It is suitable for use with llama.cpp or custom runtimes that support BitNet quantization.
Quants & memory
| Quant | File | Est. memory | vs VRAM | Result | |
|---|---|---|---|---|---|
| — | 11.2 GB | 13.8 GB | Too big | Get file → |
For 16 GB VRAM, use the 11.17 GB GGUF file; memory estimate is 13.8 GB at 8 k context, so it is close to the limit. If you have 8 GB VRAM, the model will not fit. No other quantizations are provided.
What GPU do you need for qwen2.5 coder 32b bitnet 1.58b?
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: the best quant that fits is the main file (11.2 GB file, about 13.8 GB in use).
Specifications
Architecture from the GGUF header
Source
Frequently asked questions about qwen2.5 coder 32b bitnet 1.58b
How much VRAM does qwen2.5 coder 32b bitnet 1.58b need?
About 13.8 GB at its main file with an 8K context window. That is the 11.2 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 qwen2.5 coder 32b bitnet 1.58b on an 8 GB GPU?
Not at an 8K context. The smallest file needs about 13.8 GB, so you would need a larger GPU or CPU offloading.
What is the file size of qwen2.5 coder 32b bitnet 1.58b GGUF?
It depends on the quant: file is 11.2 GB.
What context length does qwen2.5 coder 32b bitnet 1.58b support?
The GGUF header lists a trained context length of 32,768 tokens (32K). The KV cache grows with the context you use: about 2.1 GB at 8K and 8.6 GB at 32K.
Where can I download qwen2.5 coder 32b bitnet 1.58b GGUF?
The files are on Hugging Face in the tzervas/qwen2.5-coder-32b-bitnet-1.58b repository. The download buttons on this page open the file on Hugging Face. ggufmodels.com does not host any model files.
What license does qwen2.5 coder 32b bitnet 1.58b 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?
No, the estimated VRAM usage is 13.8 GB at 8 k context, which exceeds 8 GB.
Is the model fully open source?
The model is released under the Apache‑2.0 license and is available on Hugging Face.
Does it support vision tasks?
No, the model does not support vision.