Expired50% confidenceFactTime unknown
GPT 5.5 leads on TerminalBench 2.0 with a score of 87%, surpassing even Anthropic's unreleased Mythos model
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
7/2/2026
First Seen
Valid until: 9/30/2026(expired)
Sources
GPT 5.5 vs Claude Code vs DeepSeek V4: Hands-On Comparison of Three Top Coding Models
bilibili量子菠萝_
Related Claims
UnverifiedGPT 5.5 achieved a score of 87% on the TerminalBench 2.0 benchmark, significantly leading and surpassing Anthropic's yet-to-be-released Mythos model.84% similarUnverifiedGPT-5.5 与 Claude Mythos 5 在 Terminal Bench 2.1 上均停留在 88% 左右81% similarPartially VerifiedOpenAI states that GPT-5.6 achieves the current state-of-the-art on Terminal Bench 2.1.80% similarUnverifiedGPT-5.2 Codex scores slightly higher than GPT-5.2 on SWBench Pro and about 2 percentage points higher on Terminal Bench.76% similarVerifiedGPT-5.4 achieved 57.7% accuracy on SWE-Bench Pro, surpassing GPT-5.3-Codex's 56.8%.76% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/57265API
curl https://kongchang.com/api/v1/knowledge/claims/57265MCP
get_claim(id=57265)