GPT-OSS 120B

At a glance

What GPT-OSS 120B does

Tool use, Streaming, Structured output, Function calling, Web browsing, Code execution, Configurable reasoning (low/med/high), MXFP4 quantization, Self-hostable, Fine-tunable on single H100

Best for

Best open-weight model for production coding, configurable reasoning for quality/speed trade-offs, fine-tunable on single H100, agentic workflows

Works well with

Conflicts & caveats

• Trae's built-in model menu may not include GPT-OSS 120B — Trae primarily ships with Claude/GPT/Gemini/DeepSeek.
• Google Antigravity is built around Gemini 3 Pro; Claude and recent GPT-5 models also work, but GPT-OSS 120B is not in the supported model list.
• Kiro's spec mode is tuned for Claude models — GPT-OSS 120B may work but spec-driven flows are best with Claude.
• OpenAI Codex (VS Code) is tied to OpenAI's GPT-5.1 Codex family — GPT-OSS 120B is not selectable inside this integration.
• Gemini CLI is locked to Google Gemini models — your LLM is GPT-OSS 120B.
• Privacy conflict: Self-hosted GPT-OSS 120B sends code to cloud @codebase (Cursor). Use local context (Continue indexing, ChromaDB, LanceDB, pgvector, Vespa self-hosted) for true privacy.
• Claude Code only works with Claude models — your LLM is GPT-OSS 120B.
• Devin uses Cognition's proprietary model internally — your selected GPT-OSS 120B will be ignored.
• OpenAI Codex (Cloud) runs GPT-5.1 Codex internally — your selected GPT-OSS 120B will be ignored.
• Jules runs Gemini models internally — your selected GPT-OSS 120B will be ignored.
• The Claude Code SDK is tuned for Claude models — GPT-OSS 120B may work via the model adapter but loses Claude-specific features (thinking, prompt caching, sub-agents).
• Lovable uses a fixed internal model — your selected GPT-OSS 120B will be ignored.
• Replit Agent 3 uses its own model selection internally — your selected GPT-OSS 120B will be ignored.