Unverified50% confidenceBenchmarkExact time
GPT-6 Sol achieved 68.8% on the software engineering benchmark DeepSweep, approximately 1.1 percentage points behind top-tier flagship models, with costs reduced by about 80%.
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
9/29/2026
First Seen
Valid until: 12/28/2026
Sources
Related Entities
Related Claims
UnverifiedGPT-5.6 Sol achieves 62.6% on the OSWorld 2.0 computer operation benchmark, surpassing Opus 4.8, while reducing output tokens by 85%.78% similarUnverifiedThe price reduction of GPT-6 Sol increases the available quota for use in Codex, which is a positive development for developers.77% similarUnverifiedGPT-5.6 Sol achieves 92.2% on the BrowseComp browsing task benchmark.74% similarUnverifiedOn performance model code construction tasks, GPT-5.6 Sol achieved a pass rate exceeding 80%.73% similarUnverifiedIn automated business process testing, GPT-6 Sol outperformed Claude Opus 5, with a per-task cost of approximately 9% of the latter.72% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/958997API
curl https://kongchang.com/api/v1/knowledge/claims/958997MCP
get_claim(id=958997)