Type: AI model
gpt-oss-20b is an open-weight model released by OpenAI in its gpt-oss line, distributed on Hugging Face for local and self-hosted use. It is the smaller of the gpt-oss pair alongside the 120B variant. GROUNDING tracks gpt-oss-20b’s benchmarks and use in open-model deployment.
Recent Updates
- 2026-07-01: InterFLOPBench benchmarks LLMs on floating-point error classification in code (cs.AI updates on arXiv.org) · arxiv.org — LLM Evals arXiv CCSD proxy · Connected Papers · Litmaps · scite · alphaXiv · CatalyzeX · DagsHub · Gotit.pub · Hugging Face · ScienceCast · CORE · arXivLabs Lisa Taldir Qwen 3 32b · Gemini 2.5 Flash Phi-4-Reasoning DeepSeek R1T2 · GPT-OSS 120B
- 2026-07-09: SmartHomeSecure uses constrained LLM repair for Home Assistant YAML errors (cs.AI updates on arXiv.org) · arxiv.org — LLM Evals home-assistant · GPT-OSS 120B · Llama-3.1-8B · Llama-3.3-70b
- 2026-07-16: GSM-Plus-BN benchmarks Bangla mathematical reasoning in LLMs (cs.CL updates on arXiv.org) · arxiv.org — LLM Evals Open Source LLMs Qwen3-32B Llama-3.1-8b-instant llama-3.3-70b-versatile Llama-4-Scout-17B-16E-Instruct · GPT-OSS 120B
FAQ
What is gpt-oss-20b?
gpt-oss-20b is an open-weight model released by OpenAI in its gpt-oss line, distributed on Hugging Face for local and self-hosted use. It is the smaller of the gpt-oss pair alongside the 120B variant. GROUNDING tracks gpt-oss-20b’s benchmarks and use in open-model deployment.
What does this page track?
Dated radar mentions, source links, related concepts, and builder-relevant context for gpt-oss-20b, collected automatically by GROUNDING.
When was gpt-oss-20b last mentioned?
gpt-oss-20b was most recently mentioned in a radar update dated 2026-07-16.
Category: Text / Language Models