# Triton Kernel Generation on KernelBench LEVEL2 1.0 Source: [Triton Kernel Generation on KernelBench LEVEL2 1.0]() Collection index: [llms.txt]() Data updated: 2026-02-26T05:45:12.150Z Evaluation of Triton kernel generation models using the KernelBench LEVEL2 1.0 dataset, measured by specific performance factors including Fast1, Fast1.2, Fast1.5, and Fast2. A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage. | Scope | Recorded value | | --- | --- | | Task | Triton kernel generation | | Dataset (including any stated split) | KernelBench LEVEL2 1.0 | | Metric | Fast1 Score | | Unit | SCALAR | | Direction | Higher is better | | Results loaded | 17 of 17 | | Coverage | All results currently returned for this benchmark | Evaluation settings may differ between results. The original papers and recorded configurations are needed to establish comparability. A shared metric name alone does not establish an identical protocol. ## Metrics - [Fast1 Score](): Higher is better; unit: SCALAR. - [Fast1.2 Score](): Higher is better; unit: SCALAR. - [Fast1.5 Score](): Higher is better; unit: SCALAR. - [Fast2 Score](): Higher is better; unit: SCALAR. ## Recorded results | Method | Fast1 Score | Result date | Report paper | | --- | --- | --- | --- | | DR. KERNEL-14B-STTS | 80.9 | Not recorded | 698575d8c97a0b3e3b5d5e92 | | DR. KERNEL-14B-STTS | 59.8 | Not recorded | 698575d8c97a0b3e3b5d5e92 | | Claude-4.5-Sonnet | 50 | Not recorded | 698575d8c97a0b3e3b5d5e92 | | DR. KERNEL-14B | 49.2 | Not recorded | 698575d8c97a0b3e3b5d5e92 | | GPT-5 | 46.7 | Not recorded | 698575d8c97a0b3e3b5d5e92 | | DR. KERNEL-8B | 46 | Not recorded | 698575d8c97a0b3e3b5d5e92 | | AutoTriton | 30.6 | Not recorded | 698575d8c97a0b3e3b5d5e92 | | GLM-4.7 | 30 | Not recorded | 698575d8c97a0b3e3b5d5e92 | | DR. KERNEL-14B | 23.5 | Not recorded | 698575d8c97a0b3e3b5d5e92 | | GPT-5 | 22.1 | Not recorded | 698575d8c97a0b3e3b5d5e92 | | DR. KERNEL-8B | 20.6 | Not recorded | 698575d8c97a0b3e3b5d5e92 | | Claude-4.5-Sonnet | 20.5 | Not recorded | 698575d8c97a0b3e3b5d5e92 | | Qwen3-32B | 14 | Not recorded | 698575d8c97a0b3e3b5d5e92 | | Qwen3-8B | 13 | Not recorded | 698575d8c97a0b3e3b5d5e92 | | Qwen3-Coder-A30BA3 | 12.6 | Not recorded | 698575d8c97a0b3e3b5d5e92 | | Deepseek-V3.2-Thinking | 11 | Not recorded | 698575d8c97a0b3e3b5d5e92 | | Cold-Start-8B | 8.8 | Not recorded | 698575d8c97a0b3e3b5d5e92 | ### DR. KERNEL-14B-STTS Fast1 Score: 80.9. Result ID: 6998ee085698665ec4529c9c. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"scaling\_method":"sequential test-time scaling (STTS)","selection\_strategy":"best turn selection"}. This record does not assert independent reproduction. ### DR. KERNEL-14B-STTS Fast1 Score: 59.8. Result ID: 6998ee085698665ec4529c9b. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"scaling\_method":"sequential test-time scaling (STTS)"}. This record does not assert independent reproduction. ### Claude-4.5-Sonnet Fast1 Score: 50. Result ID: 6998ee085698665ec4529c91. [Reporting paper]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### DR. KERNEL-14B Fast1 Score: 49.2. Result ID: 6998ee085698665ec4529c9a. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### GPT-5 Fast1 Score: 46.7. Result ID: 6998ee085698665ec4529c90. [Reporting paper]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### DR. KERNEL-8B Fast1 Score: 46. Result ID: 6998ee085698665ec4529c99. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### AutoTriton Fast1 Score: 30.6. Result ID: 6998ee085698665ec4529c97. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### GLM-4.7 Fast1 Score: 30. Result ID: 6998ee085698665ec4529c93. [Reporting paper]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### DR. KERNEL-14B Fast1 Score: 23.5. Result ID: 6998ee085698665ec4529c7e. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Evaluation Engine":"torch.compile"}. This record does not assert independent reproduction. ### GPT-5 Fast1 Score: 22.1. Result ID: 6998ee085698665ec4529c7b. [Reporting paper]() Recorded evaluation configuration: {"Evaluation Engine":"torch.compile"}. This record does not assert independent reproduction. ### DR. KERNEL-8B Fast1 Score: 20.6. Result ID: 6998ee085698665ec4529c7d. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Evaluation Engine":"torch.compile"}. This record does not assert independent reproduction. ### Claude-4.5-Sonnet Fast1 Score: 20.5. Result ID: 6998ee085698665ec4529c7c. [Reporting paper]() Recorded evaluation configuration: {"Evaluation Engine":"torch.compile"}. This record does not assert independent reproduction. ### Qwen3-32B Fast1 Score: 14. Result ID: 6998ee085698665ec4529c95. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### Qwen3-8B Fast1 Score: 13. Result ID: 6998ee085698665ec4529c94. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### Qwen3-Coder-A30BA3 Fast1 Score: 12.6. Result ID: 6998ee085698665ec4529c96. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### Deepseek-V3.2-Thinking Fast1 Score: 11. Result ID: 6998ee085698665ec4529c92. [Reporting paper]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### Cold-Start-8B Fast1 Score: 8.8. Result ID: 6998ee085698665ec4529c98. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"training":"cold-start data"}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.