UTC

GPT OSS 20B

OpenAI · open weights · served by 56 providers · reasoning

openai/gpt-oss-20b

Weights
Open
Released
2025-08-05
Knowledge cutoff
2024-06
Context window
128K
Deprecation dates

When each platform deprecates GPT OSS 20B, as LiteLLM records it. Plan the move before the date: after it, the platform may stop serving the model or stop updating it.

  • Together AIdeprecated 2026-09-15together_ai/openai/gpt-oss-20bsuccessor: together_ai/Qwen/Qwen3.5-9B
Every upcoming deprecation →
Hugging Face repository

openai/gpt-oss-20b at commit 6cee5e81ee, last changed 2025-08-26

Licence
apache-2.0
Access
Open download
Library
transformers
Formats
.bin .safetensors
  • 1 weight file uses a pickle-based format, which can run code when loaded. Load the safetensors files instead.
  • No model signature file is published with the weights.

What the Hub's scanners report

5 weight files, 5 with a scan result. No scanner reports an issue with them.

The model at this commit, its 5 weight files with their SHA-256.

Facts from the Hugging Face Hub and its file scanners (Protect AI, Cisco, Hugging Face picklescan, VirusTotal, JFrog), as of 2026-11-18. ShieldMap does not copy model cards.

Who serves it
  • AI/ML APIopenai/gpt-oss-20b
  • Amazon Bedrockopenai.gpt-oss-20b
  • Amazon Bedrockopenai.gpt-oss-20b-1:0
  • Amazon Bedrockus-gov.openai.gpt-oss-20b-1:0
  • Clarifaiopenai/chat-completion/models/gpt-oss-20b
  • Cloudflare Workers AI@cf/openai/gpt-oss-20b
  • Cortecsgpt-oss-20b
  • Databricksdatabricks-gpt-oss-20b
  • Deep Infraopenai/gpt-oss-20b
  • DigitalOceanopenai-gpt-oss-20b
  • Eden AIcloudflare/@cf/openai/gpt-oss-20b
  • Eden AIdatabricks/databricks-gpt-oss-20b
  • Eden AIdatabricks/databricks-gpt-oss-20b@eu
  • Eden AIdeepinfra/openai/gpt-oss-20b
  • Eden AIflexai/gpt-oss-20b
  • Eden AIflexai/gpt-oss-20b@eu
  • Eden AIgreenference/gpt-oss-20b
  • Eden AIgroq/openai/gpt-oss-20b
  • Eden AIovhcloud/gpt-oss-20b
  • FastRouteropenai/gpt-oss-20b
  • FrogBotgpt-oss-20b
  • Groqopenai/gpt-oss-20b
  • Heliconegpt-oss-20b
  • Hugging Faceopenai/gpt-oss-20b
32 more providers

Impossibl (fireworks/gpt-oss-20b), Impossibl (groq/gpt-oss-20b), InferAll (openai/gpt-oss-20b), Inference APIs (openai/gpt-oss-20b), IO.NET (openai/gpt-oss-20b), Kenari (gpt-oss-20b), Kilo Gateway (openai/gpt-oss-20b), DevPass (LLM Gateway) (gpt-oss-20b), LLM Gateway (consensusprotocol/gpt-oss-20b), LLM Gateway (groq/gpt-oss-20b), LMStudio (openai/gpt-oss-20b), LowRouter (auto/openai/gpt-oss-20b), Merge Gateway (openai/gpt-oss-20b), NanoGPT (openai/gpt-oss-20b), Neon (gpt-oss-20b), Nvidia (openai/gpt-oss-20b), OCI Generative AI (openai.gpt-oss-20b), Ollama Cloud (gpt-oss:20b), OpenRouter (openai/gpt-oss-20b), Opper (gpt-oss-20b), OVHcloud AI Endpoints (gpt-oss-20b), Powerspring (gpt-oss-20b), Prime Inference (openai/gpt-oss-20b), Qiniu (gpt-oss-20b), QVAC (gpt-oss-20b), Regolo AI (gpt-oss-20b), SiliconFlow (openai/gpt-oss-20b), SOKKAN Router (openai/gpt-oss-20b), STACKIT (openai/gpt-oss-20b), Thinking Machines (openai/gpt-oss-20b), Vercel AI Gateway (openai/gpt-oss-20b), CoreWeave (openai/gpt-oss-20b)

Running it yourself

Open weights are loaded and served by code that has its own advisories. These inference servers and model loaders are on ShieldMap; load weights only from the publisher's own repository, and prefer safetensors over formats that can run code.

How supply-chain attacks on models work →

Sources: models.dev (MIT) and the LiteLLM model catalogue (MIT). Deprecation dates are as each platform published them and as LiteLLM records them; check the platform before you rely on one. Data as of 2026-11-22.