Update README.md
Browse files
README.md
CHANGED
|
@@ -38,3 +38,31 @@ This is a high-quality F16 GGUF conversion of the GPT-OSS 20B model, optimized f
|
|
| 38 |
```bash
|
| 39 |
# Download the model
|
| 40 |
huggingface-cli download davidfred/gpt-oss-20b-f16-gguf gpt-oss-20B-F16.gguf
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 38 |
```bash
|
| 39 |
# Download the model
|
| 40 |
huggingface-cli download davidfred/gpt-oss-20b-f16-gguf gpt-oss-20B-F16.gguf
|
| 41 |
+
|
| 42 |
+
|
| 43 |
+
## Hardware Requirements
|
| 44 |
+
|
| 45 |
+
- **Minimum RAM**: 16 GB (recommended: 24 GB+)
|
| 46 |
+
- **CPU**: Multi-core recommended (tested on 8 vCPU)
|
| 47 |
+
- **Storage**: ~13 GB free space
|
| 48 |
+
- **OS**: Compatible with llama.cpp (Linux, Windows, macOS)
|
| 49 |
+
|
| 50 |
+
## Performance Notes
|
| 51 |
+
|
| 52 |
+
- Efficiently runs on CPU-only setups
|
| 53 |
+
- Utilizes mixture of experts for optimal parameter efficiency
|
| 54 |
+
- Supports both interactive and batch inference modes
|
| 55 |
+
- Compatible with llama.cpp server mode for API access
|
| 56 |
+
|
| 57 |
+
## Model Origin
|
| 58 |
+
|
| 59 |
+
Converted from the original GPT-OSS 20B model using llama.cpp conversion tools. This F16 GGUF preserves all model capabilities while providing efficient inference performance.
|
| 60 |
+
|
| 61 |
+
## License
|
| 62 |
+
|
| 63 |
+
Apache 2.0 - Same as the original GPT-OSS model.
|
| 64 |
+
EOF
|
| 65 |
+
|
| 66 |
+
# Upload the README
|
| 67 |
+
huggingface-cli upload davidfred/gpt-oss-20b-f16-gguf /tmp/README.md README.md
|
| 68 |
+
|