Unverified50% confidenceFactExact time

GPT-OSS is embedded into the Codex CLI via the --oss flag, with the 20B version running on consumer hardware and the 120B version being server-grade.

1

Sources

50%

Confidence

Medium-term (~90 days)

Relevance

10/13/2026

First Seen

Valid until: 1/11/2027

Sources

Can a Local LLM Replace Claude Code? SWE-bench Is the Wrong Benchmark

Related Entities

GPT-OSSmodel Codex CLIproduct OpenAIcompany

Related Claims

Unverified单卡RTX 4090即可运行GPT-OSS 20B，单卡A100即可运行GPT-OSS 120B77% similar UnverifiedGPT-OSS 20B可在16GB显存的消费级显卡上运行75% similar UnverifiedGPT-OSS 120B需要大于80GB显存（如H100 GPU）才能运行68% similar UnverifiedGPT-OSS 20B在RTX 3090上可达约40 Token/s的生成速度，在RTX 5090上能达到200 Token/s67% similar

Cite This Claim

Stable URI

https://kongchang.com/claim/1013113

API

curl https://kongchang.com/api/v1/knowledge/claims/1013113

MCP

get_claim(id=1013113)