SWE-bench
Input—per 1M tokens
Output—per 1M tokens
Context—tokens
WeightsClosed
About
SWE-bench is ranked #16 of 29 in LLM evaluation tools on Inferse. It runs on Web, macOS, Linux.
Compared on LLM evaluation tools
- Free plan
- Yesswebench.com
- Deployment options
- bothswebench.com
Company
- Founded
- 2023swebench.com · 28 Sept 2026
Best SWE-bench alternatives
See all 12
7.4 Braintrust $249/mo first paid tier Free plan
7.4 Confident AI $200/mo first paid tier Free plan
7.4 Galileo $100/mo first paid tier Free plan
7.4 Giskard Free free plan, no paid price published Free plan
7.4 Maxim AI $29/mo first paid tier Free plan 7.4 Parea AI $150/mo first paid tier Free plan Where it ranks on Inferse
- Best LLM Evaluation Tools in 2026#16 of 29
Sources
- swebench.com· checked 28 Sept 2026




