kldzj gpt oss 120b heretic v2 needs about 71.5 GB of memory at MXFP4_MOE (a 63.4 GB file) with an 8K context window, which is more than a 48 GB GPU holds. This repository has 3 quants, from 0.2 GB to 65.4 GB (BF16).
kldzj gpt oss 120b heretic v2 – 117B parameter GPT‑OSS model
kldzj gpt oss 120b heretic v2 is a 117B parameter GPT‑OSS model released under Apache‑2.0. It is available in GGUF format with several quantizations. The model is designed for reasoning and agentic tasks and supports the harmony response format.
Good at
- Reasoning and chain‑of‑thought
- Agentic function calling
- Fine‑tuning for custom tasks
Limits
- No vision support
- Requires 8k context for memory estimates
- Large memory footprint for full precision
Best for
- Developers building reasoning agents
- Researchers fine‑tuning large language models
- Enterprise deployments needing open‑weight models
Who it’s for. This model is intended for developers and researchers who need a large, open‑weight GPT‑OSS model for advanced reasoning and agentic applications. It is suitable for those who can manage the memory requirements and are comfortable with the harmony response format.
Quants & memory
| Quant | File | Est. memory | vs VRAM | Result | |
|---|---|---|---|---|---|
| — | 0.2 GB | 0.7 GB | Fits | Get file → | |
| MXFP4_MOE | 63.4 GB (2 parts) | 71.5 GB | Too big | Get file → | |
| BF16 | 65.4 GB (2 parts) | 73.7 GB | Too big | Get file → |
For 8 GB VRAM, the only available quant is the unknown quant with 0.16 GB file size and 0.7 GB memory at 8k context. For 16 GB VRAM, use the MXFP4_MOE quant (63.39 GB file, 71.5 GB memory). For 80 GB VRAM, the BF16 quant (65.37 GB file, 73.7 GB memory) is suitable. Estimates are approximate.
What GPU do you need for kldzj gpt oss 120b heretic v2?
These figures add the file size, an fp16 KV cache at an 8K context window 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 the main file (0.2 GB file, about 0.7 GB in use).
- 6 GB VRAM: the best quant that fits is the main file (0.2 GB file, about 0.7 GB in use).
- 8 GB VRAM: the best quant that fits is the main file (0.2 GB file, about 0.7 GB in use).
- 12 GB VRAM: the best quant that fits is the main file (0.2 GB file, about 0.7 GB in use).
- 16 GB VRAM: the best quant that fits is the main file (0.2 GB file, about 0.7 GB in use).
- 24 GB VRAM: the best quant that fits is the main file (0.2 GB file, about 0.7 GB in use).
- 32 GB VRAM: the best quant that fits is the main file (0.2 GB file, about 0.7 GB in use).
- 48 GB VRAM: the best quant that fits is the main file (0.2 GB file, about 0.7 GB in use).
Which kldzj gpt oss 120b heretic v2 quant should you choose?
MXFP4_MOE (63.4 GB) is the middle option in this repository. Lower bit counts mean smaller files and more quality loss.
If memory is tight, the smallest file is the first one in the table at 0.2 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 kldzj gpt oss 120b heretic v2
How much VRAM does kldzj gpt oss 120b heretic v2 need?
About 71.5 GB at MXFP4_MOE with an 8K context window. That is the 63.4 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 kldzj gpt oss 120b heretic v2 on an 8 GB GPU?
Yes, with the smaller files: the 0.2 GB file needs about 0.7 GB at 8K context. Larger quants in this repository do not fit in 8 GB.
What is the file size of kldzj gpt oss 120b heretic v2 GGUF?
It depends on the quant: file is 0.2 GB, MXFP4_MOE is 63.4 GB, BF16 is 65.4 GB.
Where can I download kldzj gpt oss 120b heretic v2 GGUF?
The files are on Hugging Face in the bartowski/kldzj_gpt-oss-120b-heretic-v2-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 kldzj gpt oss 120b heretic v2 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?
Only the unknown quant (0.16 GB) can be loaded, but it still requires 0.7 GB memory at 8k context. It may not fit on 8 GB VRAM with other overhead.
What context length does the memory estimate assume?
The memory estimates are for an 8k context length.
Is there a vision capability?
No, the model does not support vision.