gpt-oss-safeguard-20b
OpenAIhuggingface.co/openai/gpt-oss-safeguard-20b
gpt-oss-safeguard-20b is a safety reasoning model from OpenAI built upon gpt-oss-20b. This open-weight, 21B-parameter Mixture-of-Experts (MoE) model offers lower latency for safety tasks like content classification, LLM filtering, and trust...
Updated 9 min ago · first seen 11 Sept 2026
model_01M294WW9KM7CQ555DQNXZEDRK
- Parameters
- 21.5B
- T2 · 2 h ago
- Context
- 131.1K tokens
- T2 · 2 h ago
- Released
- 18 Sept 2025
- T2 · 9 min ago
- License
- apache-2.0
- T2 · 2 h ago
Specification
- Release date
- 18 Sept 2025
Source:Hugging Face Hub (public pages, model cards, papers)T2observed 9 min agomedium
- Openness
- open-weights
Source:Hugging Face Hub (public pages, model cards, papers)T2observed 2 h agomedium
- License
- apache-2.0
Source:Hugging Face Hub (public pages, model cards, papers)T2observed 2 h agomedium
- Architecture
- GptOssForCausalLM
Source:Hugging Face Hub (public pages, model cards, papers)T2observed 2 h agomedium
- Parameters
- 21.5B
Source:Hugging Face Hub (public pages, model cards, papers)T2observed 2 h agomedium
- Context window
- 131.1K tokens
Source:OpenRouter public model & pricing listingT2observed 2 h agomedium
- Max output
- 65.5K tokens
Source:OpenRouter public model & pricing listingT2observed 2 h agomedium
- Modalities
- text
Source:OpenRouter public model & pricing listingT2observed 2 h agomedium
- Input modalities
- text
Source:OpenRouter public model & pricing listingT2observed 2 h agomedium
- Output modalities
- text
Source:OpenRouter public model & pricing listingT2observed 2 h agomedium
- Tokenizer
- GPT
Source:OpenRouter public model & pricing listingT2observed 2 h agomedium
- Base model
- openai/gpt-oss-20b
Source:Hugging Face Hub (public pages, model cards, papers)T2observed 2 h agomedium
- File size
- 13.8 GB
Source:Hugging Face Hub (public pages, model cards, papers)T2observed 2 h agomedium
- Hugging Face repo
- openai/gpt-oss-safeguard-20b
Source:OpenRouter public model & pricing listingT2observed 2 h agomedium
- Pipeline tag
- text-generation
Source:Hugging Face Hub (public pages, model cards, papers)T2observed 2 h agomedium
- Model card
Source:Hugging Face Hub (public pages, model cards, papers)T2observed 2 h agomedium
- Downloads
- 63,577
Source:Hugging Face Hub (public pages, model cards, papers)T2observed 9 min agomedium
- Likes
- 257
Source:Hugging Face Hub (public pages, model cards, papers)T2observed 9 min agomedium
- Gated
- No
Source:Hugging Face Hub (public pages, model cards, papers)T2observed 2 h agomedium
- Groq model id
- openai/gpt-oss-safeguard-20b
Source:Groq — docs & pricingT1observed 2 h agohigh
- Groq status
- preview
Source:Groq — docs & pricingT1observed 2 h agohigh
- Hf inference providers
- featherless-ai, groq
Source:Hugging Face Hub (public pages, model cards, papers)T2observed 2 h agomedium
- Last modified
- 2026-01-14T18:38:05+00:00
Source:Hugging Face Hub (public pages, model cards, papers)T2observed 2 h agomedium
- Library name
- transformers
Source:Hugging Face Hub (public pages, model cards, papers)T2observed 2 h agomedium
- Downloads all time
- 766,669
Source:Hugging Face Hub (public pages, model cards, papers)T2observed 9 min agomedium
- Model type
- gpt_oss
Source:Hugging Face Hub (public pages, model cards, papers)T2observed 2 h agomedium
- Openrouter id
- openai/gpt-oss-safeguard-20b
Source:OpenRouter public model & pricing listingT2observed 2 h agomedium
- Supported parameters
- include_reasoning, max_tokens, reasoning, response_format, seed, stop, structured_outputs, temperature, tool_choice, tools, top_p
Source:OpenRouter public model & pricing listingT2observed 2 h agomedium
- Tags
- transformers, safetensors, gpt_oss, text-generation, vllm, 8-bit, mxfp4
Source:Hugging Face Hub (public pages, model cards, papers)T2observed 2 h agomedium
- Weights dtype
- BF16, U8
Source:Hugging Face Hub (public pages, model cards, papers)T2observed 2 h agomedium
Each value shows its source, tier and observation time. Conflicting claims are kept side by side and flagged — never averaged. How AI Atlas records facts →
Provenance
Attributed facts
34
Source tiers
T1T22 / 32
Freshest observation
9 min ago
Conflicts
None
Modalities
- Modalities
- text
- Input
- text
- Output
- text
Capabilities
Tool calling
Yes
OpenRouter public model & pricing listing · T2
Structured output
Yes
OpenRouter public model & pricing listing · T2
Reasoning
Yes
OpenRouter public model & pricing listing · T2
Vision
Unavailable
Audio
Unavailable
Fine-tuning available
Unavailable
- Context window
- 131.1K tokens
Source:OpenRouter public model & pricing listingT2observed 2 h agomedium
- Max output
- 65.5K tokens
Source:OpenRouter public model & pricing listingT2observed 2 h agomedium
- Tokenizer
- GPT
Source:OpenRouter public model & pricing listingT2observed 2 h agomedium
No benchmark results recorded
Current prices
| Provider | Input / 1M | Output / 1M | Cached in | Batch in / out | Context | Observed | Source |
|---|---|---|---|---|---|---|---|
| GroqCloudopenai/gpt-oss-safeguard-20b | $0.075 | $0.30 | — | — | 131.1K | 40 min ago | console.groq.comT1 |
| OpenAI APIopenai/gpt-oss-safeguard-20b | $0.075 | $0.30 | $0.037 | — | 131.1K | 54 min ago | openrouter.aiT2 |
USD per 1M tokens as published by each provider (USD). Rows are append-only: every change is kept in the history below.
Available through 2
Price history
Input price · USD / 1M tokens
Output price · USD / 1M tokens
- OpenAI API
- GroqCloud
Memory need = bytes per parameter (4-bit ≈ 0.5 × 1.15 overhead, 8-bit 1.0, fp16 2.0) + a KV-cache allowance. Not a measurement.
| Hardware | Quantization | Memory | Est. need | Fits |
|---|---|---|---|---|
| NVIDIA DGX B200 | 4bit | 1,440 GB | 12.9 GB | Yes |
| Apple M3 Ultra | 4bit | — | 12.9 GB | Yes |
| Mac Studio (Apple M5 Ultra) | 4bit | — | 12.9 GB | Yes |
| AMD Instinct MI325X | 4bit | 256 GB | 12.9 GB | Yes |
| AMD Instinct MI300X | 4bit | 192 GB | 12.9 GB | Yes |
| Apple M2 Ultra | 4bit | — | 12.9 GB | Yes |
| NVIDIA B200 | 4bit | 180 GB | 12.9 GB | Yes |
| NVIDIA H200 | 4bit | 141 GB | 12.9 GB | Yes |
| NVIDIA H200 NVL | 4bit | 141 GB | 12.9 GB | Yes |
| Apple M1 Ultra | 4bit | — | 12.9 GB | Yes |
| Apple M3 Max | 4bit | — | 12.9 GB | Yes |
| Apple M4 Max | 4bit | — | 12.9 GB | Yes |
| Mac Studio (Apple M5 Max) | 4bit | — | 12.9 GB | Yes |
| MacBook Pro (Apple M5 Max) | 4bit | — | 12.9 GB | Yes |
| NVIDIA DGX Spark | 4bit | 128 GB | 12.9 GB | Yes |
| Apple M2 Max | 4bit | — | 12.9 GB | Yes |
| NVIDIA H100 NVL | 4bit | 94 GB | 12.9 GB | Yes |
| NVIDIA A100 80GB | 4bit | 80 GB | 12.9 GB | Yes |
| NVIDIA H100 SXM | 4bit | 80 GB | 12.9 GB | Yes |
| Apple M1 Max | 4bit | — | 12.9 GB | Yes |
| Apple M4 Pro | 4bit | — | 12.9 GB | Yes |
| Mac mini (Apple M5 Pro) | 4bit | — | 12.9 GB | Yes |
| MacBook Pro (Apple M5 Pro) | 4bit | — | 12.9 GB | Yes |
| Apple M3 Pro | 4bit | — | 12.9 GB | Yes |
| Apple M1 Pro | 4bit | — | 12.9 GB | Yes |
| Apple M2 Pro | 4bit | — | 12.9 GB | Yes |
| Apple M4 | 4bit | — | 12.9 GB | Yes |
| iMac (Apple M4) | 4bit | — | 12.9 GB | Yes |
| Mac mini (Apple M6) | 4bit | — | 12.9 GB | Yes |
| MacBook Air (Apple M5) | 4bit | — | 12.9 GB | Yes |
| MacBook Pro (Apple M5) | 4bit | — | 12.9 GB | Yes |
| NVIDIA GeForce RTX 5090 | 4bit | 32 GB | 12.9 GB | Yes |
| Apple M2 | 4bit | — | 12.9 GB | Yes |
| Apple M3 | 4bit | — | 12.9 GB | Yes |
| NVIDIA GeForce RTX 3090 | 4bit | 24 GB | 12.9 GB | Yes |
| NVIDIA GeForce RTX 4090 | 4bit | 24 GB | 12.9 GB | Yes |
| Apple M1 | 4bit | — | 12.9 GB | Yes |
Ancestors 1
- gpt-oss-20b20.9B
This model
gpt-oss-safeguard-20b
21.5B params
Quantizations 0
None recorded.
Descendants 0
None recorded.
Papers 1
- arXiv:2508.10925Active35
Repositories 0
No repositories linked yet.
As of
Rewind the record: see this entity's attributes exactly as AI Atlas knew them on a given day.
Claim history · Architecture
Architecturearchitecture1
| Value | Valid from → to | Status | Source | Confidence | Extractor |
|---|---|---|---|---|---|
| GptOssForCausalLM | → current | current | Hugging Face Hub (public pages, model cards, papers)T2 | medium | deterministic |
Claims are temporal and append-only: a new observation closes the previous claim (valid_to) instead of overwriting it. Conflicting claims from different sources are kept side by side and flagged — never averaged. Methodology →
gpt-oss-safeguard-20b: release date changed from 2025-09-18 to 2025-10-29
Release date18 Sept 2025→29 Oct 2025openroutergpt-oss-safeguard-20b: release date changed from 2025-10-29 to 2025-09-18
Release date29 Oct 2025→18 Sept 2025huggingfaceGroqCloud lists gpt-oss-safeguard-20b at $0.075 in / $0.3 out per 1M tokens
groq_pricingOpenAI API lists gpt-oss-safeguard-20b at $0.075 in / $0.3 out per 1M tokens
openrouter
| Source | Document | Type | Tier | Last observed | Snapshots |
|---|---|---|---|---|---|
| Groq — docs & pricing | console.groq.com/docs/models | model_docs | T1· Official | 40 min ago | 2 |
| Hugging Face Hub (public pages, model cards, papers) | huggingface.co/openai/gpt-oss-safeguard-20b | model_page | T2· Quality secondary | 9 min ago | 2 |
| Hugging Face Hub (public pages, model cards, papers) | huggingface.co/models?author=openai&p=0&sort=downloads | listing | T2· Quality secondary | 39 min ago | 3 |
| OpenRouter public model & pricing listing | openrouter.ai/api/v1/models | catalogue | T2· Quality secondary | 54 min ago | 2 |
| Hugging Face Hub (public pages, model cards, papers) | huggingface.co/openai/gpt-oss-safeguard-20b/raw/main/README.md | model_card | T2· Quality secondary | 1 h ago | 1 |
Tier 1 = official/primary, 2 = quality secondary, 3 = community, 4 = unverified. Every snapshot is archived; see all sources and the methodology.