gpt-oss-20b-GGUF
- https://huggingface.co/openai/gpt-oss-20b - https://huggingface.co/RedHatAI/gpt-oss-20b-speculator.eagle3
Params
20.0 B
Context
131,072
Downloads 30d
231 K
Likes
184
Download history
daily snapshots · 13 days231 K166 K
Oct 13Oct 17Oct 21Oct 25
Can you run it?
Estimated VRAM at 8K context unless noted. Pick your hardware to see the verdict per quantization.
| File | Quant | Size | Est. VRAM | Verdict on RTX 4090 · 24 GB |
|---|---|---|---|---|
| eagle3-gpt-oss-20b-Q8_0.gguf | Q8_0 | 0.9 GB | 4.5 GB | ✅ Runs comfortably |
| gpt-oss-20b-MXFP4.gguf | GGUF | 12.1 GB | 16.8 GB | ✅ Runs comfortably |
Estimate: file size × 1.1 + KV cache at 8K + 0.5 GB overhead. Not a benchmark — how we calculate this.
Run it
copy-paste, exact tags checked against the Hub$ ollama run gpt-oss-20b-gguf-ggml-org # pin the quantization explicitly $ ollama run gpt-oss-20b-gguf-ggml-org-gguf
$ huggingface-cli download ggml-org/gpt-oss-20b-GGUF-GGUF \
gpt-oss-20b-MXFP4.gguf --local-dir .
$ llama-cli -m gpt-oss-20b-MXFP4.gguf \
-c 8192 -ngl 99 -t 8 --color
$ curl -s https://aimodelscomparison.com/api/v1/models/gpt-oss-20b-gguf-ggml-org
{
"hf_id": "ggml-org/gpt-oss-20b-GGUF",
"params_b": 20.00,
"context_length": 131072,
"license": { "id": "apache-2.0", "commercial": "yes" },
"downloads_30d": 230876,
"vram_estimates": [
{ "quant": "GGUF", "gb": 16.8 },
{ "quant": "Q8_0", "gb": 4.5 }
],
"updated_at": "2026-10-20T01:00:13Z"
}
Specifications
- Parameters
- 20.0 B
- Context length
- 131,072
- Licence
- apache-2.0
- First seen on the Hub
- 2025-08-02
- Base model
- gpt-oss-20b
- Training datasets
- undisclosed
- Added to our catalog
- 2026-10-13
Compare with any text-generation model