gpt-oss-120b GGUF size and VRAM requirements
ggml-org/gpt-oss-120b-GGUF is a very large language model with 116.83 billion parameters, built on the gpt-oss architecture. It is released under the apache-2.0 license and has been downloaded 103,668 times.
gpt-oss-120b-GGUF is a Mixture-of-Experts model with 128 experts, of which 4 are active on each token. Routing only 4 of 128 experts makes it noticeably faster than a dense model of the same size — but every expert still has to be held in memory, so the numbers below are set by the full parameter count, not the active one. It is built from 36 transformer layers, a hidden size of 2,880, 64 attention heads. It uses a hybrid attention design: only 18 of its 36 layers keep a full KV cache, while the rest use linear attention with a constant-size state. Long contexts therefore cost far less memory than on a conventional model of this size.
To run ggml-org/gpt-oss-120b-GGUF locally at a 4,096-token context, its quantized versions need between 1.73 GB (Q8_0, lowest quality) and 59.97 GB (MXFP4, highest quality) of memory, weights plus KV cache and a system margin included. Context length drives that memory directly: at 4,096 tokens the KV cache for gpt-oss-120b-GGUF is about 0.14 GB, rising to roughly 4.5 GB at its full 131,072-token context. Shorter prompts free up memory for a higher-quality quantization.
For most users the best balance is BF16, needing about 2.42 GB. That means ggml-org/gpt-oss-120b-GGUF fits entirely in the VRAM of a 6 GB GPU or larger, running fully on the GPU.
Available GGUF quantizations for ggml-org/gpt-oss-120b-GGUF include Q8_0, BF16, MXFP4. The model supports a native context length of up to 131,072 tokens; a longer context grows the KV cache and the memory needed.
GGUF file size and memory by quantization
Compare real GGUF weight sizes, estimated KV cache and total memory for Q4, Q5, Q8 and every quantization published in this repository.
| Quant. | Bits | Quality | Weights | KV | Total | Speed~ | Verdict |
|---|---|---|---|---|---|---|---|
| Q8_0 | 0.06 | Very low | 0.79 GB | 0.14 GB | 1.73 GB | 505.8 t/s | Fits in VRAM |
| BF16 | 0.11 | Very low | 1.48 GB | 0.14 GB | 2.42 GB | 270.7 t/s | Fits in VRAM |
| MXFP4 | 4.34 | Good | 59.03 GB | 0.14 GB | 59.97 GB | — | Insufficient |
KV cache computed from the model's exact architecture. Speed is a rough estimate bounded by memory bandwidth.
Frequently asked questions
What kind of model is ggml-org/gpt-oss-120b-GGUF?
ggml-org/gpt-oss-120b-GGUF is a language model with 116.83 billion parameters, based on the gpt-oss architecture. It is released under the apache-2.0 license and distributed as GGUF files for local inference.
Is ggml-org/gpt-oss-120b-GGUF a Mixture-of-Experts (MoE) model?
Yes. ggml-org/gpt-oss-120b-GGUF is a Mixture-of-Experts model with 128 experts, of which 4 are activated per token. That makes it faster than a dense model of the same size, but all 128 experts must be loaded into memory, so the VRAM/RAM it needs is driven by the total parameter count, not the active one.
How does ggml-org/gpt-oss-120b-GGUF handle long context?
ggml-org/gpt-oss-120b-GGUF uses a hybrid attention design: only 18 of its 36 layers keep a full KV cache, and the rest use linear attention with a constant-size state. Long contexts cost far less memory than on a conventional model of this size.
How much VRAM do you need to run ggml-org/gpt-oss-120b-GGUF?
You need about 2.42 GB of VRAM to run ggml-org/gpt-oss-120b-GGUF entirely on the GPU using the BF16 quantization (at a 4,096-token context). Smaller quantizations lower the requirement at the cost of quality.
Can I run ggml-org/gpt-oss-120b-GGUF on an 8 GB GPU?
Yes. With 8 GB of VRAM you can run ggml-org/gpt-oss-120b-GGUF fully on the GPU using BF16 (about 2.42 GB).
Can I run ggml-org/gpt-oss-120b-GGUF on a 16 GB GPU?
Yes. With 16 GB of VRAM you can run ggml-org/gpt-oss-120b-GGUF fully on the GPU using BF16 (about 2.42 GB).
Can I run ggml-org/gpt-oss-120b-GGUF on a 24 GB GPU?
Yes. With 24 GB of VRAM you can run ggml-org/gpt-oss-120b-GGUF fully on the GPU using BF16 (about 2.42 GB).
What context length does ggml-org/gpt-oss-120b-GGUF support?
ggml-org/gpt-oss-120b-GGUF supports a native context length of up to 131,072 tokens. A longer context grows the KV cache, so it increases the memory needed to run the model.
What is the best quantization for ggml-org/gpt-oss-120b-GGUF?
For ggml-org/gpt-oss-120b-GGUF, higher bits-per-weight means better quality. A common sweet spot is a Q4_K_M or Q5_K_M quantization, which keeps most of the quality while roughly halving the memory versus 8-bit. Pick the highest quantization that still fits in your VRAM.