Expired75% confidenceFactTime unknown
The performance of GPT-OSS 120B is comparable to OpenAI O4 Mini.
1
Sources
75%
Confidence
Medium-term (~90 days)
Relevance
6/1/2026
First Seen
Valid until: 8/30/2026(expired)
Sources
Related Entities
Related Claims
UnverifiedGPT-OSS 20B performance falls between O3 mini and O4 mini.83% similarUnverifiedThe GPT-OSS 120B version activates only 5.1B parameters during inference and achieves performance close to O4-mini.80% similarVerifiedOpenAI推出GPT-4o mini作为对轻量级高性能模型市场需求的回应78% similarUnverifiedThe GPT-o3s used in the comparison test is a 20B-parameter model open-sourced by OpenAI, deployed locally via Ollama and running on 6 MXP cards.75% similarUnverifiedGPT-5.5's response speed is noticeably faster than Opus 4.7, and the team attributes this to OpenAI's hardware advantage.75% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/23262API
curl https://kongchang.com/api/v1/knowledge/claims/23262MCP
get_claim(id=23262)