Expired70% confidenceFactTime unknown
DevStral performs exceptionally well on code benchmarks such as SWE-bench.
1
Sources
70%
Confidence
Medium-term (~90 days)
Relevance
6/1/2026
First Seen
Valid until: 8/30/2026(expired)
Sources
Mistral Vibe实测：月费$15的Claude Code平替到底值不值？
bilibili安得广厦千万间678
Related Entities
Related Claims
UnverifiedSWE-Bench主要依赖单元测试是否通过来判定任务成功与否，但通过测试并不等于代码质量过关63% similarUnverifiedSWE-bench is considered by the industry to be the most practically valuable code benchmark due to its closeness to real engineering scenarios.63% similarUnverified评估智能体编程能力的主流基准包括SWE-bench、SWE-bench Verified和HumanEval63% similarUnverified传统代码评测基准（如HumanEval、MBPP）主要测试模型在孤立函数级别的生成准确率62% similarUnverifiedIn actual testing, CodeRabbit was able to accurately detect security issues involving hardcoded API keys, with review results returned almost instantaneously.61% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/20714API
curl https://kongchang.com/api/v1/knowledge/claims/20714MCP
get_claim(id=20714)