Leaderboard / OpenAI
gpt-oss-safeguard-20b
OpenAI · gpt-oss · released 2025-10-29 · data checked 2026-11-01
gpt-oss-safeguard-20b is a 20.9B-parameter Mixture-of-Experts model (3.6B active per token) from OpenAI (United States), released 2025-10-29 under the Apache-2.0 license. It doesn't have enough reported benchmarks for an index score yet. It needs about 13.8 GB of memory at 4-bit with an 8K-token context.
Safety classifier that follows your own written policy. Not a general assistant.
Benchmarks
Each bar shows this model's score. The tick marks the best medium model on that benchmark.
No benchmark results for this release yet. Know of one? Send it.
Specifications
- Lab
- OpenAI (US · United States)
- Architecture
- Mixture-of-Experts
- Total parameters
- 20.9B
- Active parameters
- 3.6B
- Size class
- Medium
- Context window
- 131K tokens
- Input modalities
- text
- Reasoning
- Reasoning (thinking)
- License
- Apache-2.0
- Openness
- Open weights
- Memory to run
- ~13.8 GB at 4-bit, 8K context (more sizes)
- Hugging Face
- 19.5K downloads (30 days) · 287 likes
Compare with
- gpt-oss-safeguard-20b vs gpt-oss-20b (20.9B, index 64.0)
- gpt-oss-safeguard-20b vs Tri-21B Think (20.7B, index 43.6)
- gpt-oss-safeguard-20b vs EuroLLM 22B Instruct (22.6B, index 19.2)
Sources
Something wrong here? Report it.
Run it locally
Uses unsloth/gpt-oss-safeguard-20b-GGUF (Q4_K_M). Swap the tag for another size: F16, Q2_K, Q2_K_L, Q3_K_M, Q3_K_S, Q4_0, Q4_1, Q4_K_M….
ollama run hf.co/unsloth/gpt-oss-safeguard-20b-GGUF:Q4_K_Mllama-server -hf unsloth/gpt-oss-safeguard-20b-GGUF:Q4_K_Mpip install -U mlx-lm
mlx_lm.chat --model lmstudio-community/gpt-oss-safeguard-20b-MLX-MXFP4vllm serve openai/gpt-oss-safeguard-20bMemory needed: weights, context cache and runtime buffers.
| Quantization | 4K context | 32K context | 131K context |
|---|---|---|---|
| 16-bit F16 | 43 GB | 44 GB | 46 GB |
| 8-bit Q8_0 | 23 GB | 24 GB | 26 GB |
| 4-bit Q4_K_M | 14 GB | 14 GB | 17 GB |