Benchmarks
Coding benchmarks
These benchmarks compare models on programming tasks. Some score the code itself. Others ask people to choose the better answer.
We track 299 benchmarks in this topic. Main benchmarks are listed by how many models they cover, most first. Where a benchmark has separate results for a subset of its tasks, those results appear right under it. Click one for its leaderboard.
Showing 201–299 of 299 benchmarks.
- SWE-Lancer4 models
- ZClawBench4 models
- CL-bench3 models
- CWE-Bench3 models
- FullStackBench en3 models
- FullStackBench zh3 models
- Gorilla Benchmark API Bench3 models
- Kimi Code Bench v23 models
- RecreationBench3 models
- CFEval2 models
- Codegolf v2.22 models
- CursorBench v3.22 models
- CyBench2 models
- DeepSWE 1.02 models
- Design2Code2 models
- FrontierCS2 models
- HumanEval-Mul2 models
- Image2FloorPlan2 models
- MBPP EvalPlus2 models
- MiMo Cyber Bench2 models
- MiMo Visual Coding2 models
- PatchBench2 models
- QwenSVG2 models
- SecureSWE2 models
- SeedClawBench2 models
- Trae Code Gen2 models
- Vibe Code Bench2 models
- APEX-SWE1 model
- BigCodeBench-Full1 model
- BigCodeBench-Hard1 model
- CADGenBench1 model
- CathedralBench1 model
- CC-Bench-V2 Backend1 model
- CC-Bench-V2 Frontend1 model
- CC-Bench-V2 Repo Exploration1 model
- CL-bench (Life)1 model
- CRUXEval-Input-CoT1 model
- CRUXEval-Output-CoT1 model
- CVE-Bench1 model
- CyberSecEval 41 model
- DS-Arena-Code1 model
- Flame-VLM-Code1 model
- FrontierSWE (Impl.)1 model
- HumanEval-ER1 model
- Instruct HumanEval1 model
- Kernel Bench L31 model
- LiveCodeBench(01-09)1 model
- LongCodeBench1 model
- MBPP ++ base version1 model
- MBPP EvalPlus (base)1 model
- OctoCodingBench1 model
- QwenClawBench1 model
- Surge AI Coding Quality1 model
- SWE-MM1 model
- SWE-Perf1 model
- SWE-Review1 model
- SWT-Bench1 model
- Terminus1 model
- VIBE1 model
- VIBE Android1 model
- VIBE Backend1 model
- VIBE iOS1 model
- VIBE Simulation1 model
- VIBE Web1 model