Unverified50% confidenceBenchmarkExact time
In the T3 code improvement task, GPT-6.1 Sol scored 87.4, beating Astra (83.8) and Grok 4.7 (80.7), at a cost of only $2.15.
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
10/1/2026
First Seen
Valid until: 12/30/2026
Sources
Related Entities
Related Claims
Partially VerifiedThe standard API input and output token price for GPT-6.1 Sol is only one-fifth that of Astra.79% similarUnverifiedGPT-6.1 Sol achieves the highest score among all models with a per-task cost below $0.30.76% similarUnverifiedIn the DeepSWE 1.1 benchmark, GPT-6.1 Sol scored 75.2%, Astra scored 74.8%, Cloud Sonnet 5.5 scored 71.0%, and GPT-6 Sol scored a top of 68.8%.72% similarUnverifiedSol is the most capable of the three models, priced on par with GPT-5.5, and is suitable for complex coding and long-task scenarios.72% similarUnverifiedGPT-5.6 Sol achieves 92.2% on the BrowseComp browsing task benchmark.71% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/964502API
curl https://kongchang.com/api/v1/knowledge/claims/964502MCP
get_claim(id=964502)