Compare GPT-4.1 and GPT OSS 120B, both from OpenAI, in 3 community votes, gpt oss 120b wins 100% of head-to-head duels, context windows of 1.0M vs 131K, tested across 54 shared challenges.
Pick GPT OSS 120B. In 3 blind votes, GPT OSS 120B wins 100% of the time. That's not luck.
GPT OSS 120B particularly excels in Reasoning. GPT OSS 120B is 10x cheaper per token — worth considering if cost matters.
GPT OSS 120B is cheaper on both — 11× input, 10× output
GPT-4.1 uses 3.3x more lists
Compare GPT-4.1 and GPT OSS 120B, both from OpenAI, in 3 community votes, gpt oss 120b wins 100% of head-to-head duels, context windows of 1.0M vs 131K, tested across 54 shared challenges.
Pick GPT OSS 120B. In 3 blind votes, GPT OSS 120B wins 100% of the time. That's not luck.
GPT OSS 120B particularly excels in Reasoning. GPT OSS 120B is 10x cheaper per token — worth considering if cost matters.
GPT OSS 120B is cheaper on both — 11× input, 10× output
GPT-4.1 uses 3.3x more lists