Verified65% confidenceFactTime unknown

In early 2024, the strongest AI models scored only in the 10%–20% range on SWE-Bench.

3

Sources

65%

Confidence

Long-term

Relevance

6/1/2026

First Seen

Sources

Gemini 3 Flash In-Depth Review: Comprehensive Testing of Coding, Multimodal, and Writing Capabilities

bilibilikate人不错

Related Entities

SWE-Benchconcept

Related Claims

Unverified在SWE-bench Full版本上，多数模型的通过率仍低于20%75% similar UnverifiedIn early 2023, mainstream models' SWE-bench scores were generally below 5%; by mid-2025, top models had surpassed 70%.74% similar UnverifiedAcceptance rates at top AI conferences are often below 20%.70% similar

Cite This Claim

Stable URI

https://kongchang.com/claim/19989

API

curl https://kongchang.com/api/v1/knowledge/claims/19989

MCP

get_claim(id=19989)